University of Southern California Dissertations and Theses

Loss of checkpoint controls in acute lymphoblastic leukemia

(USC Thesis Other)

Loss of checkpoint controls in acute lymphoblastic leukemia

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content LOSS OF CHECKPOINT CONTROLS IN ACUTE LYMPHOBLASTIC LEUKEMIA

by

Srividya Swaminathan

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GENETIC, MOLECULAR AND CELLULAR BIOLOGY)

May 2013

Copyright 2013 Srividya Swaminathan

ii
Dedication

I dedicate this work to my parents and grandmother who have supported me in all my
academic endeavors. Without their constant encouragement, it would have not been
possible for me to pursue my PhD.

iii
Acknowledgements

I hereby take the opportunity to thank all those people who have helped me during the
course of my PhD.
First and foremost, I would like to thank my advisor Dr. Markus Müschen for his
guidance and support during the entire time span of my PhD. Dr. Müschen has given me
the opportunity to carry out cutting-edge research in Immunology. My research in his lab
has not only enabled me to expand my scientific knowledge, but has also helped me to
hone my technical and presentation skills. Through his mentorship, I have learnt to
appreciate the relevance of basic immunology research in a disease setting.
Next, I would like to thank all my co-workers in the lab (past and current) who have
made the entire experience of PhD not only rewarding but enjoyable. I would like to
extend my special gratitude to Dr. Soo-Mi Kweon, Lars Klemm, Carina Ng and Eugene
Park who have made direct contribution to some of the experiments described in this
thesis. I would also like to thank our collaborators namely, Dr. Ari Melnick, Dr. Cheryl
Willman, Dr. Mel Greaves, Dr. Kazuhiko Igarashi, Dr. Andrew Hall and Dr. Rafael
Casellas for providing us mouse models and patient data related to the work described
in this thesis.
I would like to thank my committee members- Dr. Michael Lieber, Dr. Nora Heisterkamp,
Dr. Jae Jung and Dr. Yong-Mi Kim for being extremely supportive. I thank them for their
guidance and for helping me analyze scientific questions from different dimensions.
Finally, I would like to thank my family and friends both in India and in US, for
encouraging and supporting me throughout this process. Most importantly, I would like to
thank my parents who have been instrumental in motivating me to achieve my academic
goals. They have always taught me to strive harder and face any challenges with
optimism. None of this would have been possible without their support.
iv
Table of contents
Dedication ....................................................................................................................... ii

Acknowledgements ........................................................................................................ iii

List of Tables ................................................................................................................. vii

List of Figures ............................................................................................................... viii

Abstract ......................................................................................................................... xii

Chapter 1. B cell development and pre-B acute lymphoblastic leukemia (ALL) ............... 1
1.1. Hematopoesis ...................................................................................................... 1
1.2. Early B cell development ..................................................................................... 2
1.3. Allelic exclusion during Ig recombination ............................................................. 5
1.4. Pre-B ALL- A result of defect in early B cell development .................................... 6
1.5. DNA Recombination at Ig loci and leukemogenesis ............................................. 8
1.6. Pre-B cell receptor (Pre-BCR) signaling ..............................................................10
1.7. Checkpoints during normal B cell development ..................................................13
1.8. An overview of the objectives and conclusions of this thesis. ..............................16

Chapter 2. Cooperation between AID and RAG1/ RAG2 V(D)J recombinase in the
clonal evolution of childhood ALL. .................................................................................19
2.1. Clonal evolution of childhood ALL .......................................................................19
2.2. AID and clonal evolution of B cell malignancies ..................................................19
2.3. Cooperation between AID and RAG enzymes in the acquisition of genetic
alterations in pre-B ALL .............................................................................................20
2.4. Requirement of a second hit in TEL-AML1-driven childhood ALL .......................21
2.5. AID and RAG1/2 as factors inducing the second hit ............................................22
2.6. Infection or inflammation as a trigger for AID expression in pre-B cells ...............23
2.7. Checkpoints protect early B cells from genetic instability ....................................23

Chapter 3. IL7R as a guardian against leukemogenesis at the pre-BCR checkpoint ......25
3.1. Introduction .........................................................................................................25
3.2. Materials and Methods........................................................................................26
3.2.1. Extraction of bone marrow cells from mice ...................................................26
3.2.2. Mice strains and bonemarrow culture ...........................................................26
3.2.3. Retrovirus production and transduction ........................................................27
3.2.4. Quantitative RT-PCR ...................................................................................27
3.2.5. Western blotting ...........................................................................................28
3.2.6. Flow cytometry and Cell cycle analysis ........................................................28
3.3. Results ...............................................................................................................29
3.3.1. Link between AID and pre-B ALL .................................................................29
3.3.2. Early murine B cells are safeguarded from pre-mature AID activation ..........29
3.3.3. Identification of IL7R signaling as the molecular mechanism safeguarding
murine pre-B cells from pre-mature AID activation .................................................30
3.3.4. IL7R signaling safeguards human pre-B cells from pre-mature AID
expression .............................................................................................................32
v
3.3.5. Identification of JAK-STAT and PI3K signaling pathways as negative
regulators of pre-mature AID expression ...............................................................33
3.3.6. Fraction D pre-B cells respond to inflammatory signals ................................36
3.3.7. Fraction D – The subset most vulnerable to leukemogenesis.......................39
3.3.8. Pre-mature AID activation promotes the clonal evolution of a B cell clone in
the bone marrow of a child carrying the TEL-AML1 rearrangement. ......................41
3.3.9. Genetic ablation of Aid and Rag1 abrogates leukemia initiation upon
inflammation ..........................................................................................................42
3.3.10. Repeated exposure of Fraction D cells to inflammation results in genetic
instability ................................................................................................................44
3.4. Conclusions ........................................................................................................47
3.5. Limitations and future perspectives .....................................................................48

Chapter 4. Diverse roles of BACH2 in B-lymphoid cells- Relevance to B-ALL ...............50
4.1. Location and structure of BACH2 ........................................................................50
4.2. BACH2 is a B-lymphoid transcription factor ........................................................51
4.3. BACH2 can act as a transcriptional repressor and activator ...............................52
4.4. BACH2 triggers oxidative stress-induced cell death ............................................53
4.5. Bach2 is a common integration site for viruses ...................................................54
4.6. BACH2 is a putative tumor suppressor in B cell lymphoma .................................56
4.7. The BACH2 and BCR-ABL1 connection .............................................................57

Chapter 5. BACH2 is required for negative selection at the pre-BCR checkpoint ...........59
5.1. Introduction .........................................................................................................59
5.2. Materials and Methods........................................................................................60
5.2.1. Extraction of bone marrow cells from mice ...................................................60
5.2.2. Bach2
+/+
and Bach2
-/-
mice ...........................................................................60
5.2.3. In vitro differentiation assay using BCR-ABL1 Tyrosine Kinase Inhibitors
(TKIs) .....................................................................................................................60
5.2.4. Flow cytometry .............................................................................................61
5.2.5. Clonality analysis and spectratyping of B cell populations ............................61
5.2.6. Retroviral transduction .................................................................................61
5.2.7. Cloning of MSCV Bach2-ER
T2
IRES GFP vector ..........................................62
5.2.8. Sequence analysis of V
H
-DJ
H
gene rearrangements ....................................62
5.2.9. Assay to measure RAG1/RAG2 recombinase activity ..................................62
5.2.10. Western Blotting .........................................................................................63
5.2.11. Growth competition assay ..........................................................................63
5.3. Results ...............................................................................................................64
5.3.1. BACH2 induces negative selection of pre-B cells .........................................64
5.3.2. Involvement of PAX5, ARF and TP53 (p53) in BACH2-induced negative
selection ................................................................................................................67
5.3.3. BACH2 regulates the level and activity of the RAG enzymes .......................69
5.3.4. BACH2 and BCL6 maintain balance between negative selection and survival
of early B cells .......................................................................................................71
5.4. Conclusions ........................................................................................................73
5.5. Limitations and future perspectives .....................................................................74

Chapter 6. BACH2 protects against leukemogenesis at the the pre-BCR checkpoint ....75
6.1. Introduction .........................................................................................................75
6.2.1. Patient samples, human cells and cell lines .................................................76
vi
6.2.2. Extraction of bone marrow cells from mice ...................................................76
6.2.3. In vivo model for Myc driven leukemia .........................................................76
6.2.4. Bach2
+/+
, Bach2
-/-
,

Bcl6
+/+
and Bcl6
-/-
mice....................................................76
6.2.5. BCR-ABL1 Tyrosine Kinase Inhibitors (TKIs) ...............................................77
6.2.6. Western blotting ...........................................................................................77
6.2.7. Flow cytometry .............................................................................................77
6.2.8. Colony forming assay ..................................................................................77
6.2.9. Retroviral transduction .................................................................................78
6.2.10. Cell-cycle analysis .....................................................................................78
6.2.11. Quantitative single-locus ChIP ...................................................................78
6.2.12. PCR amplification and sequencing of BACH2 coding region ......................78
6.2.13. Sequence alignment to identify mutations ..................................................79
6.2.14. ROS staining ..............................................................................................79
6.2.15. Affymetrix GeneChip analysis ....................................................................79
6.2.16. BACH2 gene expression data and clinical outcome ...................................80
6.3. Results ...............................................................................................................81
6.3.1. Transcriptional inactivation of BACH2 in ALL ...............................................81
6.3.2. BACH2 inhibits oncogenic transformation of pre-B cells ...............................88
6.3.3. Mechanisms of BACH2- induced tumor suppression ...................................95
6.3.4. BACH2-induced apoptosis is dependent on p53 status .............................. 103
6.3.5. BACH2 as an effector in PAX5-mediated tumor suppression in ALL .......... 107
6.4. Conclusions ...................................................................................................... 108
6.5. Limitations and future perspectives ................................................................... 109

Chapter 7. Discussion ................................................................................................. 111

Bibliography ................................................................................................................ 117

Appendices ................................................................................................................. 128
Appendix A: Supplementary information for Chapter 3 ............................................ 128
Appendix B: Supplementary information for Chapter 5 ............................................ 132
Appendix C: Supplementary information for Chapter 6 ............................................ 135

vii
List of Tables

Table 1.1. Stages of early B cell development (Hardy’s Fractions). ................................ 5
Table 3.1. AID-dependent mutations in TEL-AML1 pre-B ALL. ......................................46
Table A1. Sequences of oligonucleotide primers for qRT-PCR used in chapter 3 ........ 131
Table B1. List of oligonucleotide primers for chapter 5 ................................................ 134
Table C1. Sequences of oligonucleotide primers used in chapter 6. ............................ 137
Table C2. Somatic mutations of the BACH2 gene in patient-derived Ph
+
ALL cells. ..... 139

viii
List of Figures

Figure 1.1. Flow chart depicting the process of hematopoesis. ...................................... 2

Figure 1.2. Steps in early B cell development and their corresponding surface marker
expressions. ................................................................................................................... 5

Figure 1.3. Pre-B ALL results from malignant transformation of B-cell precursors. ......... 8

Figure 1.4. Digrammatic representation of DNA recombination at Ig loci. ......................10

Figure 1.5. Pre-B cell receptor signaling pathway. .........................................................12

Figure 1.6. Key checkpoints during early B cell development. .......................................15

Figure 2.1. Mechanistic basis of clustering of chromosomal translocation breakpoints at
CpG sites. .....................................................................................................................21

Figure 2.2. Representation of the structure of TEL-AML1 fusion gene. ..........................22

Figure 3.1. Correlation between hypermutation targets of AID and childhood ALL. .......29

Figure 3.2. Early B cells are safeguarded from pre-mature AID activation until Fraction D.
......................................................................................................................................30

Figure 3.3. Pre-B cell receptor signaling upregulates AID at Fraction D by
downregulating surface expression of IL7Rα. ................................................................31

Figure 3.4. In vitro differentiation of early B cells from Fraction C’ to D by IL7 withdrawal
upregulates AID. ............................................................................................................32

Figure 3.5. IL7R signaling safeguards human pre-B cells from premature AID expression
before Fraction D. ..........................................................................................................33

Figure 3.6. STAT5 is a negative regulator of AID expression. ........................................34

Figure 3.7. FoxO factors and PTEN are transcriptional activators of Aid at Fraction D. 35

Figure 3.8. Fraction D cells from AID-GFP reporter mice respond to a surrogate of
infection like LPS by upregulating AID. ..........................................................................37

Figure 3.9. Fraction D cells from AID Cre-YFP reporter mice respond to inflammatory
signals from LPS by upregulating AID. ..........................................................................38

Figure 3.10. Fraction D cells lack protective mechanisms against pre-mature AID
activation. ......................................................................................................................39

Figure 3.11. Rag1/Rag2 recombinases are upregulated by the PTEN/FoxO pathway
when cells differentiate from Fraction C’ to D. ................................................................40

ix
Figure 3.12. AID overexpression accelerates TEL-AML1-driven leukemia in vivo. .........41

Figure 3.13. AID and RAG1 are required for the leukemic transformation of TEL-AML1
carrying pre-B cell clones in the context of inflammation. ..............................................43

Figure 3.14. mSKY reveals clonal architecture of TEL-AML1 pre-B ALL. .......................45

Figure 4.1. Positions of MLV integration within the Bach2 locus. ..................................55

Figure 5.1. Absence of Bach2 impairs negative selection process. ................................65

Figure 5.2. Bach2 is both required and sufficient for negative selection of B cells lacking
functional IgH rearrangements. .....................................................................................66

Figure 5.3. PAX5 is upstream of BACH2-induced negative selection of B cells with non-
functional IgH rearrangements. .....................................................................................68

Figure 5.4. BACH2 regulates level of Rag1/ Rag2 V(D)J recombinase. .........................69

Figure 5.5. BACH2 regulates activity of RAG1/ RAG2 enzymes. ...................................70

Figure 5.6. Loss of Bach2 impairs Vκ-Jκ light chain rearrangement. ..............................70

Figure 5.7. BACH2 and BCL6 maintain balance between negative selection and survival
of early B cells. ..............................................................................................................72

Figure 6.1. Human Ph
+
ALLs are characterized by low BACH2 expression levels. ........82

Figure 6.2. B-ALL patients show decreased BACH2 expression levels upon relapse. ...83

Figure 6.3. Deletions on chromosome 6 involving BACH2 locus in relapse ALL patients.
......................................................................................................................................83

Figure 6.4. High BACH2 expression levels are an indicator of favorable outcome in B-
ALL patients. .................................................................................................................86

Figure 6.5. Pediatric ALL patients with low BACH2 expression levels have poor overall
prognosis. ......................................................................................................................87

Figure 6.7. Bach2 is required for MYC-induced apoptosis. ............................................90

Figure 6.8. Bach2 prevents leukemic transformation by Myc in vitro. .............................90

Figure 6.9. Bach2 prevents leukemic transformation by Myc in vivo. .............................92

Figure 6.10. Bach2 prevents leukemic transformation by Stat5
CA
. .................................93

Figure 6.11. BACH2 limits permissiveness of pre-B cells to Myc expression levels and to
ROS accumulation.........................................................................................................94

Figure 6.12. Loss of Bach2 during progressive leukemic transformation. ......................95
x

Figure 6.13. Bach2 inhibits leukemic transformation by activating the classical Arf-
Mdm2-p53 tumor suppressor pathway...........................................................................96

Figure 6.14. Phenotype of Bach2-deficient acute lymphoblastic leukemia cells. ............97

Figure 6.15. BACH2 reverses BCL6-dependent gene expression changes. ..................98

Figure 6.16. Bach2 and BCL6 maintain balance between oncogene-induced apoptosis
and leukemic transformation of early B cells. .................................................................99

Figure 6.17. BACH2 and BCL6 share binding sites on promoters of well-known tumor
suppressor genes. ....................................................................................................... 100

Figure 6.18. BACH2 inhibits recruitment of BCL6 to CDKN2A (Arf) and TP53 (p53)
promoters. ................................................................................................................... 101

Figure 6.19. BCL6 reverses BACH2-mediated transcriptional activation of Arf/p53. .... 102

Figure 6.20. Relapse free survival probability of pediatric ALL patients is dependent on
the relative levels of BACH2 and BCL6. ...................................................................... 103

Figure 6.21. Bach2-induced tumor suppression is p53-dependent. ............................. 104

Figure 6.22. Overexpression of Bach2 in primary Ph
+
ALL cells induces drastic cell death
by a p53-dependent mechanism. ................................................................................. 105

Figure 6.23. Induction of BACH2 expression is compromised in ALLs containing PAX5
fusions. ........................................................................................................................ 107

Figure 7.1. IL7R signaling and BACH2 represent two failsafe barriers against
leukemogenesis at the pre-BCR checkpoint. ............................................................... 115

Figure A.1. Infectious and inflammatory origins of childhood pre-B ALL. ..................... 128

Figure A.2. IL7R signaling protects against leukemogenesis at the pre-BCR checkpoint
by preventing genetic instability. .................................................................................. 128

Figure A.3. IL7R signaling prevents LPS responsiveness by blocking AID expression.
.................................................................................................................................... 129

Figure A.4. Mouse spectral karyotyping (mSKY) reveals aneuploidy and chromosomal
translocations in TEL-AML1 pre-B ALL. ....................................................................... 129

Figure B.1. Sequence analysis of V
H
(D)J
H
junctions in Bach2
+/+
and Bach2
-/-
bone
marrow and splenic B cells. ......................................................................................... 132

Figure B.2. Inducible overexpression of Bach2 aids the clearance of pre-B cells with
non-productive V
H
(D)J
H
rearrangements...................................................................... 132

Figure B.3. Bach2
-/-
pre-B cells display reduced expression of Bach2 mRNA. ............. 133
xi

Figure B.4. Overexpression of Bach2 triggers expression of Rag1/ Rag2 recombinase in
pre-B cells. .................................................................................................................. 134

Figure C.1. BACH2 is increased in Ph
+
ALL cells upon TKI treatment. ........................ 135

Figure C.2. BACH2 is affected by somatic mutations in a fraction of Ph
+
ALL cases. .. 135

Figure C.3. Bach2 prevents leukemic transformation by Myc in vivo. .......................... 136

Figure C.4. Btg2 is downregulated upon loss of Bach2. ............................................... 136

xii

Abstract

Leukemogenesis is a multi-step process which involves the disruption of the normal
development of a hematopoetic cell. Precursor-B acute lymphoblastic leukemia (Pre-B
ALL) is one such example of the process of leukemogenesis that results from
deregulated early B cell development in the bone marrow. Typically, checkpoints which
are present at every stage of development safeguard an early B cell from leukemic
transformation. This thesis focuses on two such critical cell signaling mechanisms that
act at the pre-B cell receptor (pre-BCR) checkpoint (Fraction C’), and thereby protect the
cell from overt leukemogenesis. While one of the mechanisms protects the B cell from
acquiring genetic alterations, the other eliminates deleterious B cells by a process known
as negative selection.
In the first half of this thesis, we highlight the novel role of IL7R as the guardian of the B
cell genome before Fraction D. We show that IL7R carries out this function by preventing
pre-mature expression of AID and deleterious rise in the levels of the RAG1/RAG2
recombinases. Both these enzymes are notorious in causing genetic instability and have
been previously implicated in the process of leukemogenesis. However, the natural
safeguard mechanisms that prevent their concomitant expression have not been
elucidated. Our study provides novel insight in this direction. In addition, we shed light on
the role of inflammatory cues in accelerating the process of leukemogenesis, when the
safeguard is lost or compromised. We demonstrate that this study is particularly relevant
to the clonal evolution of certain childhood leukemia like the TEL-AML1 subgroup, which
require second hits for full-blown leukemogenesis to occur.
In the latter half of this thesis, we focus on another mechanism of leukemic
transformation that occurs at the pre-BCR checkpoint. We show that the natural process
of negative selection (apoptosis) at the pre-BCR checkpoint is crucial to thwart
xiii
leukemogenesis. We identify the BTB domain containing B-lymphoid protein BACH2 as
the key inducer of the negative selection process. Role of BACH2 has not been well-
characterized in early B cells. Therefore, our study is the first of its kind to highlight the
role of BACH2 in negative selection. By extrapolating our understanding of BACH2 in B
cell development to pre-B ALL, we demonstrate that this protein possesses leukemia-
suppressive properties. Therefore, BACH2 represents another paradigm of safeguard
against leukemogenesis at the pre-BCR checkpoint.
Through our studies on both IL7R and BACH2, we highlight the importance of a tightly
controlled early B cell development process that safeguards the cell from leukemic
transformation. Such an understanding of leukemogenesis is required for identifying
better treatment strategies for ALL patients.

1
Chapter 1. B cell development and pre-B acute lymphoblastic leukemia
(ALL)

Acute Lymphoblastic Leukemia (ALL) is a malignant overproduction of immature
lymphocytes in the bone marrow. It arises when lymphocytes are arrested during their
early development. It is the most common form of childhood cancer and comprises 23%
of all childhood cancers (Pui et al., 2004). Data from the National Cancer Institute state
that there are 4000 cases of ALL diagnosed in the United States every year of which,
3000 are children.

1.1. Hematopoesis
B lymphocytes are a class of immune cells which arise from the hematopoetic stem cells
(HSCs) by a process of differentiation called hematopoesis (Figure 1.1). The
development of B cells occurs in the bone marrow. Once the development process is
complete, the mature B cells migrate to the spleen and peripheral lymphoid organs.
There are multiple steps starting from a hematopoetic stem cell which ultimately lead to
the formation of a mature B cell. The fate of cells at every stage of hematopoesis is
determined by interplay of various transcription factors and cues from the bone-marrow
micro-environment. HSCs differentiate into either common myeloid progenitors (CMPs)
or common lymphoid progenitors (CLPs). The fate of a hematopoetic stem cell, that is,
whether it gives rise to a CMP or CLP at any given point of time is decided by the factors
Ikaros and Pu.1 (Georgopoulos et al., 1994; Wang et al., 1996; Reynaud et al., 2008).
Higher levels of the former cause a hematopoetic stem cell to become a CLP (DeKoter
and Singh, 2000). The common lymphoid progenitors can give rise to B-cells, T-cells,
Natural Killer (NK) cells and dendritic cells (DC) (Galy et al., 1995).
Differentiation of CLPs into B cells is induced by transcription factors E2A and EBF
(Busslinger et al., 2000). These factors induce PAX5 which is necessary for the
2
restriction of lymphoid progenitors to the B-lineage (Nutt et al., 1999). Throughout this
thesis, we focus on B lymphocytes and how deregulated cell signaling during their early
development leads to precursor-B ALL.
Figure 1.1. Flow chart depicting the process of hematopoesis.
The blood stem cell (HSC) is the parent cell that gives rise to all the hematopoetic
lineages. (Adapted from National Cancer Institute website)

1.2. Early B cell development
The process of forming the heavy and light chains of the antibody (B cell receptor/
immunoglobulin/ Ig) on the surface of a B cell occurs through several steps which
involve DNA recombination. This entire process is referred to as early B cell
differentiation. The steps involved in this process are called Hardy’s Fractions (Hardy
and Hayakawa, 2001).
3
The very early stage B cells which arise from the common lymphoid progenitor carry no
rearrangement of either heavy or light chain loci and are referred to as pro-B or pre-pro-
B cells (Fraction A; Table 1.1). These cells are AA4.1
+
B220
+
CD43
+
CD25
-
(Figure 1.2;
Rolink et al., 1999). One of the molecules that specifies B cell fate at the common
lymphoid progenitor to pre-proB stage is the cytokine interleukin-7 (IL7) (Kikuchi et al.,
2008). IL7 binds to IL7 receptor (IL7R) which signals through the JAK-STAT signaling
pathway and allows proliferation and survival of early B cells. IL7R signaling is a pre-
requisite for murine B cell and T cell development, and in its absence, mice lack B and T
lymphocytes (Sudo et al., 1993). In the case of humans, IL7R is not required for
development of B cells. However, it is an absolute requirement for T cell development
and loss of its function results in severe combined immunodeficiency (SCID) (Plum et al.,
1996).
The differentiation of B cells is triggered by the rearrangement of the heavy chain by a
process known as V(D)J recombination (Section 1.5). The recombination is mediated by
the Recombination Activating Genes (RAG1 and RAG2) which create double stranded
breaks (DSBs) in the DNA and join them by Non-Homologous End Joining (NHEJ)
(Schatz et al., 1989; Grawunder et al., 1998; Lieber et al., 2004; Lieber MR, 2010).
During the heavy chain rearrangement, the cells first recombine the D
H
and J
H
segments
to form a D
H
J
H
joint. This is followed by the joining of the V
H
segment to the rearranged
D
H
J
H
joint to form the variable region of the heavy chain (Schatz et al., 1989; Gellert M,
2002; Igarashi et al., 2002). The constant region of the Ig heavy chain (IgH) is linked
shortly after this by alternative splicing.
The cells rearranging the Ig heavy chain comprise Fraction B/C and are referred to as
Pro-B or Pre-B-I cells (Table 1.1). The Fractions B and C have CD19
+
CD43
+
CD24
+
phenotype

(Figure 1.2). A completely rearranged heavy chain with its variable and
constant regions is called the μ chain. The μ heavy chain couples with the surrogate light
4
chain components, namely, VpreB (VPREB1) and λ5 (IGLL1) to form the pre-B cell
receptor (pre-BCR) on the cell surface (Karsuyama et.al, 1994; Shimizu et al., 2002).
This stage of development where the pre-BCR is on the cell surface is called the large
cycling pre-BII stage (Fraction C’) (Rolink et al., 1999; Hardy and Hayakawa, 2001;
Herzog et al., 2009). The pre-BCR has dual function. On the one hand, it promotes
proliferation and survival of Fraction C’ cells, and on the other, it negatively auto-
regulates itself to induce a cell cycle arrest which subsequently triggers the
rearrangement of the light chain.
Light chain recombination involves the joining of the V
L
and J
L
segments to form the V
L
J
L

joint. The cells rearranging the Ig light chain are the small pre-B cells (Fraction D) (Hardy
and Hayakawa 2001; Rolink et al., 2000a; Rolink et al., 2000b). Fractions C’ and D
share a common surface marker phenotype CD43
low
CD19
+
CD25
+
(Rolink et al., 1999).
However, Fraction D cells are smaller in size than Fraction C’.
Differentiation of cells from Fraction C’ to D is also marked by reduction in signaling from
the IL7R. As a result of this attenuation in IL7R signaling, Fraction D cells move from an
environment in the bone marrow containing high IL7 to a region having lower levels of
IL7 (Johnson et al., 2008). Later in this thesis, we highlight the relationship between
leukemia initiation and such gradient in cytokine signaling during B cell differentiation.
Once the light chain is successfully rearranged, it couples with the heavy chain to form
the BCR (IgM/antibody) which is exported to the cell surface. This coupling involves the
replacement of the surrogate light chain by the conventional light chain to form the IgM
+

B cells. The cells which carry surface IgM are called immature cells (Fraction E) (Table
1.1; Figure 1.2). Fraction E cells then leave the bone marrow to populate peripheral
lymphoid organs as mature B cells (Fraction F; Figure 1.2).

5
Table 1.1. Stages of early B cell development (Hardy’s Fractions).
The rearrangement status of Ig loci at every fraction is shown below. (Adapted from:
Hardy and Hayakawa, 2001).

Phenotypic
subset
Fr. A Fr. B/C Fr. C' Fr. D Fr .E
Philadelphia
nomenclature
Pre-Pro-B Pro-B Early Pre-B Late Pre-B New-B
Basel
nomenclature
Pro-B Pre-B-I Pre-B-I Pre-B-II
IgH locus Germline Rearranging VDJ VDJ VDJ
IgL locus Germline Germline Germline Rearranging VJ

Figure 1.2. Steps in early B cell development and their corresponding surface
marker expressions.
Nomenclature of different stages of early B cell development and the cell surface
markers to recognize them are shown. (Adapted from Li et al., 1996).

1.3. Allelic exclusion during Ig recombination
Not all rearrangements that occur at the heavy and light chain loci are productive. For
example, a heavy chain rearrangement is considered productive only if it satisfies 2
criteria: 1) the recombination event at the heavy chain should lead to the formation of μ-
6
chain protein and, 2) there should be successful pairing between the μ-chain and the
surrogate light chain components (Rolink et al., 1999).
IgH rearrangement initially begins only at either the paternal or maternal allele. If the first
allele does not result in the successful formation of a pre-BCR, then the second allele
gets a chance to rearrange. If the recombination at the second allele results in a
successful rearrangement, then the first allele is inactivated and not used to form the
pre-BCR. This phenomenon is termed as allelic exclusion (Yount et al., 1968;
Mostoslavsky et al., 2004). If both the alleles fail to produce a functional rearrangement,
the cell is directed towards apoptosis.
A similar process of allelic exclusion occurs during light chain recombination. The light
chain possesses two classes of genes per allele (κ and λ). The κ locus is more
frequently rearranged than the λ locus. Only if both the maternal and paternal alleles at
the κ loci fail to rearrange, the cells try to recombine the λ loci (Lewis et al., 1982). This
doubles the chances of rearrangement at the light chain loci as compared to the heavy
chain. In case rearrangements by both the classes of the light chain fail, apoptosis
results (Rolink et al., 1999).

1.4. Pre-B ALL- A result of defect in early B cell development
B cell precursor acute lymphoblastic leukemia (BCP ALL) or pre-B ALL arises due to a
blockade at one of the stages of early B cell development described in Section 1.2. Such
a developmental block causes cells at the arrested stage to proliferate excessively
without differentiating, leading to leukemia. Leukemia is one of the cancers
characterized by chromosomal aberrations and aneuploidy (Raimondi et al., 1999; Look
et al., 1997). A large number of these chromosomal aberrations are in the form of
translocations, which arise from double-stranded DNA breaks (DSBs) in two different
parts of the genome, followed by their juxtaposition.
7
Some of the commonly found chromosomal rearrangements in pre-B ALL are MLL-AF4,
TEL-AML1, BCR-ABL1 and E2A-PBX1 (Matsuo et al., 1998). Of these, the MLL-AF4
translocation is found in infant leukemia (Gale et al., 1997; Kim-Rouille et al., 1999;
Heerema et al., 1999), and arises from a developmental arrest at the pro-B cell stage
(Bertrand et al., 2001).
BCR-ABL1-driven pre-B leukemia (also known as Philadelphia chromosome
+
or Ph
+

leukemia) results from t(9;22) reciprocal translocation (Nowell and Hungerford, 1960).
Ph
+
ALL is characterized by an excessive proliferation of cells at the pre-B stage of
development. BCR-ABL1 is more frequently observed in adult leukemia and represents
one of the leukemic subgroups with the worst prognosis (Faderl et al., 2000). This fusion
protein encodes a constitutively active cytoplasmic tyrosine kinase which activates a
multitude of oncogenic signaling pathways like the JAK-STAT, MAPK and PI3K signaling
pathways (Deininger et al., 2000; Sattler et al., 1997). We primarily discuss this
subgroup of leukemia in the context of a tumor suppressor gene BACH2, in chapters 4,
5 and 6 of this thesis.
TEL-AML1 (ETV6-RUNX1) is the most frequent subtype of childhood ALL with the most
favorable outcome (Romana et al., 1995a; Romana et al., 1995b; Shurtleff et al., 1995;
Raynaud et al., 1996; Borkhardt et al., 1997). Even though the TEL-AML1 translocation
is acquired in utero, overt leukemia develops in these children only much later after
acquisition of secondary genetic alterations (Ford et al., 1998). We have conducted in-
depth inquiry into the mechanism of development of overt leukemia in this subgroup in
Chapters 2 and 3 of this thesis.
Another commonly observed translocation in ALL is E2A-PBX1. This translocation is
present in childhood leukemia that arise at the pro or pre-B cell stage (Aspland et al.,
2001). A diagrammatic representation of the subgroups of leukemia discussed above is
shown in Figure 1.3.
8
Figure 1.3. Pre-B ALL results from malignant transformation of B-cell precursors.
Shown below is a classification of leukemia based on the type of chromosomal
translocation they carry, and the stage of B cell development from which they arise.

1.5. DNA Recombination at Ig loci and leukemogenesis
RAGs (RAG1/ RAG2) enzymes are essential for the process of DNA recombination to
generate BCR and TCR by B and T lymphocytes, respectively. As discussed previously
in this chapter, the variable region of the heavy chain is formed by a process known as
V(D)J recombination. The joining of Variable, Diversity and Joining segments is
abbreviated as V(D)J recombination.
The Recombination Signal Sequences (RSS) adjacent to all V, D and J segments are
the start sites for VDJ recombination. RSS sequences consist of a conserved heptamer
and nonamer sequence separated by 12 or 23 bp of non conserved spacer DNA
sequences (Figure 1.4). The length of these spacer regions is crucial as recombination
9
occurs between RSS regions with only 12 and 23 bp spacers (Figure 1.4; Tonegawa et
al., 1983).
The B-lymphoid specific proteins RAG1 and RAG2 recognize the RSS sequences and
ensure the correct 12/23 pairing. The first step is the induction of double strand breaks.
A nick is made at the 5’ end of the heptamer leaving a 5’ phosphoryl group on the RSS
and a 3’ hydroxyl at the coding end. Next, the 3’ hydroxyl is joined to the 5’ phosphoryl
group at the same nucleotide position on the opposite strand resulting in a DNA hairpin
(Gellert M, 2002). Following this, the Ku70/Ku80 proteins are recruited to these double
strand breaks followed by the DNA-PK
CS
and Artemis (Blier et al., 1993; Ma et al., 2002;
Lieber MR, 2010). The DNA hairpins are recognized by Artemis and cleaved by single
strand cleavage (Ma et al., 2002). Nucleotides can then be added (P nucleotides) or
removed from the generated coding ends. Addition of non-templated nucleotides by
Terminal deoxynucleotidyl Transferase (TdT) increases the diversity of the Ig.
The open ends of the coding joints or the signal joint are joined by Non-Homologous End
Joining (NHEJ) machinery involving DNA ligase IV and XRCC4 (Gu et al., 2007a; Gu et
al., 2007b; Lieber MR, 2010). The usual arrangement of RSS sequences is such that the
joined coding segment remains in the chromosome while the junction of RSSs (a signal
joint) is excised as a circular DNA which is later lost from the cells (Figure 1.4C, Gellert
M, 2002).
As described above, the normal process of DNA rearrangement to generate functional
immunoglobulins induces double-stranded breaks and thus makes the pre-B cell more
prone to acquisition of genetic alterations. This would explain why a large proportion of
leukemia are characterized by chromosomal aberrations. In this thesis, we provide a
mechanistic explanation of how processes operating during normal B cell development
(like DNA recombination) can increase susceptibility to leukemia. We also discuss in
10
these chapters the circumstances under which RAG1 and RAG2 enzymes lead to
leukemic transformation.
Figure 1.4. Digrammatic representation of DNA recombination at Ig loci.
A: The consensus heptamer and nonamer sequences of RSS are shown with the spacer
which is either 12 or 23 nucleotides long. B: Arrangement of RSSs at the heavy and light
chains of Ig loci. A 12 spacer RSS is indicated by an open triangle and a 23 spacer RSS
by a black triangle. C: Recombination results in the formation of a coding joint and signal
joint. RSSs are denoted by triangles as in Figure 1.4B and their coding flanks are
denoted by rectangles. (Adapted from Gellert M, 2002).

1.6. Pre-B cell receptor (Pre-BCR) signaling
The pre-BCR is formed by the coupling of μ heavy chain with surrogate light chain
components VpreB (VPREB1) and λ5 (IGLL1). The pre-BCR is present on the surface of
the cell at Fraction C’. The extracellular part of the pre-BCR is linked to two signaling
chains referred to as Igα and Igβ, which contain an Immunoreceptor Tyrosine-based
Activation Motif (ITAM). The pre-BCR serves as an immunological synapse where the
Igα and Igβ signaling chains serve as docking sites to assemble and activate the
components of the pre-BCR signaling cascade. The pre-BCR can signal autonomously
without a ligand.
As mentioned earlier, the pre-BCR has dual function- 1) proliferation and survival and, 2)
growth arrest and differentiation. Therefore, it serves as a checkpoint (Section 1.7) in
11
early B cell development which decides whether the cell should survive for further
differentiation or undergo apoptosis. If the pre-BCR does not carry a functional heavy
chain rearrangement, it is eliminated by apoptosis. This phenomenon is termed as
negative selection. If the IgH rearrangement is functional, the cell survives and the pre-
BCR signals for proliferation and differentiation as described below.
One of the main proteins that dock at the immunological synapse and mediate the dual
function of pre-BCR is Spleen Tyrosine (Y) Kinase (SYK). SYK induces proliferation and
survival by activating a number of downstream survival pathways, like the PI3K signaling
pathway (Figure 1.5; Deane and Fruman 2004; Kanie et al., 2004; Streubel et al., 2006).
SYK also subsequently performs the opposing function of growth arrest and
differentiation by phosphorylating and activating B cell linker or adaptor (BLNK/ SLP65/
BASH) (Figure 1.5; Jumaa et al., 1999; Guo et al., 2000; Chiu et al., 2002).
Phosphorylated SLP65 recruits Bruton’s Tyrosine Kinase (BTK) (Figure 1.5) which inturn
activates PLCγ2 (PLCG2) (Hashimoto et al., 1999; Ishiai et al., 1999; Su et al., 1999;
Chiu et al., 2002). PLCγ2 generates diacylglycerol and IP3 by hydrolysis of PI(4,5)P
2
,
which serve as second messengers for activation of PKCs and Ca
2+
release from
cytoplasmic stores.
SLP65 triggers the process of κ light chain rearrangement by activating its downstream
effectors (Yamamoto et al., 2006; Herzog et al., 2008). SLP65 blocks the proliferation
signal emanating from SYK and thus shifts the cell from proliferation to differentiation
mode. The exact mechanism by which this switch occurs is still unknown. SLP65
inactivates the PI3K signaling pathway, and thereby activates the FoxO factors. The
latter transcriptionally activate the RAG enzymes and allow for Ig light chain
rearrangement (Figure 1.5; Herzog et al., 2009). Slp65
-/-
pre-B cells are arrested at
Fraction C’ (large pre-B) stage of development (Jumaa et al., 1999), and are constantly
proliferating at this stage.
12
Pre-BCR has also been implicated in inducing the mutator enzyme Activation Induced
Cytidine Deaminase (Aicda/Aid) through Btk (Han et al., 2007). This implies that pre-
BCR may mimic the signaling of a mature BCR in the context of infection or inflammation.
The role played by inflammation/ infection in inducing leukemogenesis at the Fraction C’
to D transition is the prime focus of chapters 2 and 3 of this thesis.
Figure 1.5. Pre-B cell receptor signaling pathway.
The illustration below shows all the major signaling components downstream of the pre-
BCR and how they regulate each other. Also shown are the key functions of pre-BCR in
inducing differentiation of cells from Fraction C’ to D. (Adapted from Herzog et al., 2009).

Understanding the mechanism of pre-BCR signaling was important to us for two major
reasons. Firstly, we wanted to understand how full-blown leukemia results from pre-
BCR-induced activation of the Rag enzymes and Aid. Secondly, we wanted to dissect
how signaling molecules at the pre-BCR checkpoint maintain balance between negative
selection of non-functional B cells (apoptosis) and leukemic transformation.

13
1.7. Checkpoints during normal B cell development
During early B cell development, there are multiple checkpoints which maintain the
quality of B cells generated at every stage (Figure 1.6). It is these checkpoints that
ensure that early B cells do not go astray and undergo leukemic transformation. A loss
of checkpoint controls by normal differentiation or by mutagenesis of important signaling
molecules is deleterious to a B cell and increases the probability of leukemia initiation.
The first checkpoint comes into play when multi-lineage progenitors commit to the B
lineage (Figure 1.6; Nutt et al., 1999). The guardian at this checkpoint is Paired Box
Protein 5 (PAX5). PAX5 carries out the above function by repressing B-lineage
‘inappropriate’ genes and simultaneously activating B-lineage specific genes (Cobaleda
et al., 2007). Even after commitment to B-lineage, there are multiple checkpoints which
check the quality of B cells produced at every stage of differentiation. These checkpoints
are important because they ensure that the B cell repertoire is only made up of cells with
functional immunoglobulin rearrangements that are specific to foreign antigens and do
not react to self.
The μ heavy chain of the pre-B cell receptor represents the next checkpoint, after PAX5
(Figure 1.6; Hendriks et al., 2004). This checkpoint filters out and eliminates cells which
have not successfully rearranged their heavy chain, thereby preventing the accumulation
of deleterious cells in the B cell repertoire. At this checkpoint two opposing processes
occur- one serves to eliminate pre-B cells with non-functional V(D)J rearrangements
(negative selection) and the other allows the survival and proliferation of pre-B cells
carrying functional heavy chain rearrangements. A disruption of the balance between the
two processes leads to leukemic transformation of the pre-B cell. The importance of
pre-BCR signaling in carrying out the above mentioned processes has been discussed
earlier in Section 1.6.
14
The cell signaling mechanisms that mediate the process of negative selection after
heavy chain rearrangement are not fully understood. We were thus particularly
interested in the molecular players involved in this process of negative selection at the
pre-BCR checkpoint and their relationship to leukemogenesis. In Chapters 5 and 6, we
elucidate the role of a B-lymphoid protein, BACH2, which is downstream of PAX5, in
negative selection and leukemogenesis. We have previously shown in our lab that proto-
oncogene B Cell Lymphoma 6 (BCL6) is upregulated by the pre-BCR (Duy et al., 2010).
We have also shown that BCL6 is essential for survival at the pre-BCR checkpoint to
allow light chain recombination to occur (Duy et al., 2010). In this thesis, we propose
that interplay between BACH2 and BCL6 maintains the balance between negative
selection and survival (Chapter 5). We also provide experimental evidence to
demonstrate how this balance is upset during leukemic transformation of a pre-B cell
(Chapter 6). In addition, we focus on another signaling cascade which is crucial at the
pre-BCR checkpoint for blocking leukemogenesis, namely the IL7 receptor (IL7R)
signaling (chapter 3).
The third and final checkpoint comes into the picture following the completion of negative
selection. This checkpoint is governed by one of the signaling components of pre-BCR,
SLP65 (Figure 1.6). SLP65 induces differentiation of pre-B cells that have survived and
successfully passed the pre-BCR checkpoint control. Loss of SLP65 leads to leukemic
transformation of pre-B cells by blocking their differentiation from Fraction C’ to D
(Jumaa et al., 1999; Ta et al., 2010).
In summary, disruption of checkpoints at any stage of early B cell development pre-
disposes the cell to leukemic transformation. Therefore, study of checkpoint signaling
mechanisms are crucial to understand the process of leukemogenesis.
15
Figure 1.6. Key checkpoints during early B cell development.
The molecules regulating the quality control of B cells at every stage of differentiation
are marked in red.

16
1.8. An overview of the objectives and conclusions of this thesis.
Childhood pre-B acute lymphoblastic leukemia (ALL) arises as a result of deregulated
cell signaling during early development of B cells in the bone marrow. In this thesis, we
focus on two out of the multitude of deregulated signaling pathways in leukemia. Both
the signaling pathways we discuss here reflect the importance of the pre-B cell receptor
(pre-BCR) checkpoint in preventing leukemia-initiation. Findings from this study provide
novel insight into the importance of IL7R signaling and BACH2-mediated negative
selection as safeguards against leukemogenesis at this checkpoint.
B cells represent a subset in the body which is vulnerable to accumulation of DNA level
changes, primarily due to the expression of enzymes like RAGs (chapter 1) and AID
(chapter 2). Previous work by Tsai et al. and others show that genetic alterations
acquired in pre-B ALL (particularly, chromosomal translocations) carry a distinct
signature which is characteristic of AID and RAG expression (Tsai et al., 2008). However,
signaling mechanisms at critical checkpoints during early B cell development keep the
expression of these two enzymes at bay and prevent leukemogenesis. In chapters 2 and
3 of this thesis, we describe how IL7R signaling at the pre-BCR checkpoint inhibits Aicda
and Rag1/Rag2 expression, and thereby, behaves as a guardian against accumulation
of genetic changes that cause overt leukemia.
We first highlight how the aberrant expression of AID during early B cell development is
deleterious in the sense that it allows accumulation of DNA mutations. We show that
IL7R safeguards early B cells until Fraction D from pre-mature AID expression. We then
predict the environmental cues (eg, inflammation or infection) which may enhance AID
expression to deleterious levels at Fraction D.
Mechanisms regulating the expression of RAG enzymes during early B cell development
have been well characterized. In this thesis, we focus on the cooperation of the RAG
enzymes with AID in causing double-stranded DNA breaks and ultimately, chromosomal
17
translocations. We identify Fraction D as the subset of early B cells which are most
vulnerable to leukemic transformation due to concomitant AID and RAG expression.
Finally, we lay down the mechanistic basis for the clonal evolution of childhood ALL and
the role of inflammation or infection in accelerating this process. All our findings provide
compelling evidence for IL7R expression as the guardian of the B cell genome until the
pre-BCR checkpoint control is complete.
In the second part of this thesis, we study how deregulated negative selection at the pre-
BCR checkpoint can lead to pre-B ALL. To this end, we identify BACH2 (a transcription
factor downstream of PAX5) as one of the critical molecules mediating this checkpoint
control. We illustrate the requirement of BACH2 for negative selection at the pre-BCR
checkpoint and quality control in early B cell development. We show that BACH2
eliminates non-functional B cells by inducing apoptosis. We also highlight how the
balance between BACH2 and BCL6 (a pro-survival molecule) maintains balance
between negative selection and survival at the pre-BCR checkpoint (chapter 5).
Having identified BACH2 as a pre-requisite for negative selection in early B cells, we
hypothesize that loss of BACH2 leads to loss of checkpoint control and ultimately
leukemogenesis. We thus go on to study the relationship between BACH2 and pre-B
ALL (chapter 6). BACH2 has been extensively studied in lymphoma where it is proposed
to play a tumor suppressive role. We and others have shown that BACH2 is one of the
lymphoid specific proteins that is strongly upregulated upon treatment of Ph
+
(BCR-
ABL1) leukemia with tyrosine kinase inhibitors. Collaborative studies in childhood ALL
patients identify BACH2 as a prognostic factor associated with favorable outcome.
Based on these findings and our findings from early B cell development, we predict that
BACH2 may be a putative tumor suppressor in pre-B ALL. To test our theory, we study
whether loss of Bach2 expression impacts leukemic transformation by oncogenes like
BCR-ABL1, cMyc and Stat5. We find that loss of Bach2 expression accelerates the
18
process of oncogenic transformation and allows pre-B cell survival and proliferation
(chapter 6).
We finally characterize the mechanism by which BACH2 induces the above described
tumor suppressive effect in pre-B ALL. We identify that the tumor suppression by
BACH2 is p53-dependent. We also highlight how the balance between BACH2 and
BCL6 (a pro-survival molecule) is critical to decide the fate of a pre-B cell subjected to
an oncogenic stimulus. Through our studies on BACH2, we provide yet another example
of how signaling mechanisms at the pre-BCR checkpoint safeguard early B cells from
development of ALL (chapter 6).
To sum up, we characterize two protective signaling mechanisms which inhibit
leukemogenesis till the pre-BCR checkpoint is crossed- 1) IL7R, which safeguards the
early B cell genome from DNA damage and, 2) BACH2, which by way of negative
selection (apoptosis) rids the B cell repertoire of non-functional deleterious B cells. Loss
or bypass of either one of the above mentioned safeguards will result in leukemic-
transformation of a pre-B cell.

19
Chapter 2. Cooperation between AID and RAG1/ RAG2 V(D)J recombinase
in the clonal evolution of childhood ALL.
2.1. Clonal evolution of childhood ALL
Childhood acute lymphoblastic leukemia arises from a pre-leukemic pre-B cell clone.
ALL, like other cancers follows the Darwinian mechanism of clonal evolution. Clones
evolve through the interaction of selectively advantageous ‘driver’ lesions (also called as
first hit), selectively neutral ‘passenger’ lesions, and deleterious ‘hitchhiker’ mutations
leading to full-blown leukemia (Greaves and Maley, 2012).

2.2. AID and clonal evolution of B cell malignancies
AID is a 24kDa enzyme of the APOBEC family which mutates cytosines in DNA to uracil
and methylcytosines to thymine (Figure 2.1A, Muramatsu et al., 1999). The conversion
of cytosine to uracil is a reversible process which can be corrected by a high fidelity
repair process called base excision repair (BER). However, the conversion of
methylcytosine to thiamine cannot be repaired with high fidelity as it uses an error prone
repair mechanism called mismatch repair (MMR). This leads to the retention of
mutations in the target genome.
AID is generally upregulated in mature germinal center B cells upon antigen exposure,
where it is involved in generating antibody diversity to counteract infections (Muramatsu
et al., 1999). Antibody diversity is created by somatic hypermutation and class switching
at Ig loci. Recent studies have shown that AID targets regions of the genome other than
the Ig loci (Liu et al., 2008; Staszewski et al., 2011). Liu et al. show that numerous genes
are targeted for mutation by AID but escape because of the combined action of BER and
MMR. Interestingly, a vast majority of these AID hypermutation targets are also
frequently altered in pre-B ALL (Mullighan et al., 2007). However, demonstrating the
20
presence of AID in a pre-B cell would be required to bridge the gap between AID and
pre-B leukemogenesis.
Previous studies from our lab have shown that AID is the primary driver of mutations in
BCR-ABL1 leukemia (Feldhahn et al., 2007), and that it promotes the progression of
chronic myeloid leukemia to lymphoid blast crisis (Klemm et al., 2009). AID has also
been shown to accelerate the clonal evolution in BCR-ABL1-driven ALL (Gruber et al.,
2010). There is not much known about the role of AID in other subtypes of childhood
ALL carrying the TEL-AML1, E2A-PBX1 and MLL-AF4 rearrangements.

2.3. Cooperation between AID and RAG enzymes in the acquisition of genetic
alterations in pre-B ALL
ALL involves a number of genetic alterations in the form of point mutations, deletions,
aneuploidy and chromosomal translocations (Raimondi et al., 1999; Look et al., 1997).
Chromosomal translocations occur due to double stranded breaks in DNA. In early B cell
development, DSBs result from V(D)J and VJ rearrangements of the Ig heavy and light
chains respectively, caused by RAG1 and RAG2 (described in detail in chapter 1).
RAG1 and RAG2 recognize and cleave the DNA at RSS which flank different segments
of the heavy and light chains. This is followed by the non-homologous end joining
(NHEJ) of the required segments by addition of N-nucleotides (chapter 1). Some
breakpoints in leukemia, for example, the E2A-PBX1 subtype, have some features of
V(D)J recombination but lack functional RSS (Agrawal et al., 1998; Hiom et al.,1998).
Since RAGs act as transposases in vitro, they could most likely be one of the factors in
inducing chromosomal translocations, even though the exact sequence which they
recognize in such a situation is not known.
Tsai et al. propose a mechanism by which the co-operation between AID and RAG
enzymes can lead to chromosomal translocations in leukemia. The theory they propose
21
involves the clustering of translocation hot spots at CpG sites in the genome. The
proposed mechanism involves DNA mismatches at CpG sites caused by AID converting
methylcytosines to thymine (Figure 2.1A). These sites of DNA mismatch may then be
recognized by RAG enzymes that cause DSBs leading to chromosomal translocations
(Figure 2.1B).
Figure 2.1. Mechanistic basis of clustering of chromosomal translocation
breakpoints at CpG sites.
A: AID is a deaminase that can act on either cytosine or methylcytosine as substrates
and convert them to uracil and thymine, respectively. Conversion of cytosine to uracil is
repaired by uracil glycosylases (UNG) and is highly efficient. Conversion of
methylcytosine to thymine is repaired by thymine glycosylases but this class of enzymes
is not as efficient at UNG. B: Conversion of methylcytosine to thymine by AID creates a
DNA mismatch and a bubble like structure at the site of change, which can serve as a
good substrate for RAG1/ RAG2-induced cleavage.

2.4. Requirement of a second hit in TEL-AML1-driven childhood ALL
TEL-AML1 (ETV6- RUNX1)-driven ALL is the most frequent subtype of childhood ALL
and has the most favorable outcome (Romana et al., 1995a; Romana et al. 1995b;
Shurtleff et al., 1995; Raynaud et al., 1996; Borkhardt et al., 1997). The rearrangement
involves the formation of a chimeric transcription factor, by the fusion of helix loop helix
of TEL (a member of ETS-like transcription factor family) with AML1, the DNA binding
subunit of AML1-CBFβ transcription factor complex (Figure 2.2).
22
The TEL-AML1 rearrangement is acquired in utero and is found in the Guthrie blood
spots of healthy neonates. Interestingly, even though 1 in 100 healthy newborns carry
the rearrangement, only <1% of these children develop overt leukemia. Furthermore,
studies on monozygotic twins carrying the TEL-AML1 rearrangement show only a
concordance of 10% with variability in latency of the disease between the twins
(Wiemels et al, 1999). This indicates the requirement of secondary lesions for overt
leukemogenesis. It is thus compatible with the minimal two step model for childhood
leukemia, with the chromosomal translocations acting as the ‘initiating hit’ (Greaves et al.,
2003).
Figure 2.2. Representation of the structure of TEL-AML1 fusion gene.
TEL-AML1 fusion gene results from t(12;21) translocation. It comprises of the DNA
binding domains of both TEL (Helix Loop Helix) and AML1 (Runt Homology Domain),
making it a chimeric transcription factor.

2.5. AID and RAG1/2 as factors inducing the second hit
Both AID and RAG enzymes have been shown to be involved in B cell malignancies,
with the former being more involved in mature B cell lymphomas and the latter in pre-B
leukemia. Both these enzymes are notorious in inducing genetic instability in B cells with
a specific affinity for genes involved in B cell development and gene regulation. TEL-
AML1 leukemia arises from a pre-B cell and RAG enzymes are active at this stage.
However, the concomitant expression of AID which allows acquisition of secondary
genetic lesions needs to occur at this stage in order to cause overt leukemia. Han et al.
23
have shown that AID may be expressed at very low levels in early stages of B cell
development, but the exact mechanism by which this occurs is not known.
Therefore, our study is focused on two questions- 1) what is the safeguard mechanism
in early B cells which prevents pre-mature AID expression and, 2) under which
conditions is this safeguard mechanism lost and what factor triggers a sizeable amount
of AID expression to cause a genetic lesion?

2.6. Infection or inflammation as a trigger for AID expression in pre-B cells
Infection and inflammation activate AID expression in mature germinal center B cells.
However, little is known about the effects of exposure of pre-B cells to infectious/
inflammatory agents, when their safeguard mechanism against AID is removed. Dr. Mel
Greaves and Dr. Joseph Wiemels propose that children who acquired the TEL-AML1
gene rearrangement in utero are more likely to develop leukemia in response to delayed
and recurrent infections. Epidemiological analyses have shown that exposure to
infections early in childhood have a protective effect against development of leukemia in
children (Ma et al., 2002; Perrillat et al., 2002). Based on these, we hypothesize that
exposure of a non-safeguarded TEL-AML1 pre-B cell to antigen (infection/inflammation),
may lead to overt leukemia, by upregulating AID.

2.7. Checkpoints protect early B cells from genetic instability
Checkpoint controls during early B cell development protect against leukemogenesis
(Chapter 1, Section 1.7). There are no studies to date elucidating the checkpoint
signaling mechanisms that guard the pre-B cell genome from acquisition of mutations.
Moreover, little is known about the common molecular mechanism behind the acquisition
of chromosomal translocations in leukemia. Our study (Chapter 3) shows experimentally
the central unifying factors that lead to genetic instability, a characteristic feature of B-
24
cell leukemia. We demonstrate the importance of checkpoints in preventing genetic
instability during early B cell development and thereby, in inhibiting leukemogenesis.
We show that hypermutator AID can be aberrantly activated at early stages of B cell
development, under special circumstances. Our study proves that these special
circumstances come into effect only after the pre-BCR checkpoint is crossed, in other
words, when a pre-B cell is no longer under strict surveillance.
The study also highlights the cooperation between AID and the RAG enzymes in giving
rise to a pre-leukemic clone which leads to overt leukemic transformation. Finally, we
provide the molecular and genetic basis for the “Delayed Infections Hypothesis” which
has been proposed to be one of the causative factors for leukemia progression in
children.

25
Chapter 3. IL7R as a guardian against leukemogenesis at the pre-BCR
checkpoint

3.1. Introduction
Little is known about the common molecular mechanism underlying pre-B ALL
development. In many cases, childhood ALL can be retraced to a recurrent genetic
lesion in utero which establishes a pre-leukemic clone. The TEL-AML1 fusion gene for
instance, arises prenatally (Ford et al., 1998) and defines the most frequent subtype of
childhood ALL. Strikingly, ~1 of 100 healthy newborns carry a TEL-AML1 pre-leukemic
clone, but only less than 1% of these children eventually develop leukemia. This
indicates the requirement of additional secondary mutations to give rise to overt
leukemia in these children (Greaves et al., 2003). We thus study the molecules
responsible for introducing these genetic changes and the natural safeguard
mechanisms that block them.
As discussed in Chapter 2, AID and RAG1/RAG2 have been hypothesized to cooperate
in introducing genetic lesions which lead to full blown leukemia (Tsai et al., 2008).
However, for such a cooperation to occur, both these enzymes must be expressed at the
same stage of early B cell development. As previously discussed, while AID is required
for somatic hypermutation and class switching during late stages of B cell development,
its pre-mature activation may be deleterious. Signaling mechanisms at checkpoints
during early B cell development prevent pre-mature expression of AID. Here, we show
that IL7R signaling prevents deleterious AID activation and mutation accumulation until
the clearance of the pre-BCR checkpoint.
In this chapter, we first identify the stage in early B cell development which is prone to
accumulation of mutations due to concomitant expression of both AID and RAG
enzymes. Following this, we dissect the molecular mechanism behind the activation of
both Aid and Rag1/Rag2 at this stage. We then examine the role of inflammation in
26
increasing AID expression to deleterious levels during early B cell development. Finally,
we study the implications of cooperation between AID and the RAG enzymes on the
clonal evolution of childhood ALL, using TEL-AML1 driven pre-B ALL as an example.

3.2. Materials and Methods
3.2.1. Extraction of bone marrow cells from mice
Bone marrow cells were obtained by flushing cavities of femur and tibia with PBS. After
filtration through a 40-μm filter and depletion of erythrocytes using a lysis buffer (BD
PharmLyse, BD Biosciences, San Jose, CA), washed cells were either frozen for storage
or subjected to further experiments.
3.2.2. Mice strains and bonemarrow culture
Slp65
-/-
mice: Generated in Hassan Jumaa’s Laboratory (Jumaa et al., 1999).
Stat5
fl/fl
mice: Generated in Lothar Henninghausen’s laboratory (Cui et.al., 2004).
Pten
fl/fl
bone marrow: Obtained from Hong Wu’s laboratory (Lesche et.al., 2002).
AID-GFP transgenic mice: Generated in Rafael Casellas’ laboratory (Crouch et al.,
2007).
AID-Cre YFP transgenic mice: Generated in Rafael Casellas’ laboratory (Crouch et al.,
2007).
TEL-AML1 transgenic bone marrow: Generated in Mel Greaves’ laboratory (Ford et al.,
2009).
Aid
-/-
mice: Generated in Honjo laboratory (Miramatsu et al., 2000).
Rag1
-/-
mice: Generated in Papaioannou laboratory (Mombaerts et al., 1992).
The bone marrow cells were harvested and cultured with 10 ng/ml IL-7 on either
RetroNectin- (Takara, Madison, WI)-coated non tissue culture dishes or on irradiated
OP9 stroma layer. All pre-B cells derived from bone marrow of mice were maintained in
Iscove’s modified Dulbecco’s medium (IMDM, Invitrogen, Carlsbad, CA) with GlutaMAX
27
containing 20% fetal bovine serum, 100 IU/ml penicillin, 100 μg/ml streptomycin, 50 µM
2-mercaptoethanol and 10 ng/ml recombinant mouse IL-7 (Peprotech, Rocky Hill, NJ) at
37°C in a humidified incubator with 5% CO
2
. All mouse experiments were subject to
institutional approval by Childrens Hospital Los Angeles IACUC.
3.2.3. Retrovirus production and transduction
Transfections of MSCV-based retroviral constructs were performed using Lipofectamine
2000 (Invitrogen, Carlsbad, CA) with Opti-MEM media (Invitrogen). Retroviral
supernatant was produced by co-transfecting 293FT cells with the plasmids pHIT60
(gag-pol) and pHIT123 (ecotropic env; kindly provided by Donald B Kohn, UCLA).
Cultivation was performed in high glucose Dulbecco’s modified Eagle’s medium (DMEM,
Invitrogen) with GlutaMAX containing 10% fetal bovine serum, 100 IU/ml penicillin, 100
μg/ml streptomycin, 25mM HEPES, 1 mM sodium pyruvate and 0.1 mM non-essential
amino acids. Regular media were replaced after 16 hours by growth media containing 10
mM sodium butyrate. After 8 hours incubation, the media was changed back to regular
growth media. 18 hours later, the virus supernatant was harvested, filtered through a
0.45 µm filter and loaded by centrifugation (2000 x g, 90 min at 32 °C) two times on
50 ug/ml RetroNectin (Takara, Madison, WI) coated non-tissue 6-well plates. 1-2 x 10
6

pre-B cells were transduced per well by centrifugation at 600 x g for 30 minutes and
maintained overnight at 37°C with 5% CO
2
for 2 days before transferring into culture
flasks.
3.2.4. Quantitative RT-PCR
Quantitative real-time PCR carried out with the SYBRGreenER mix from Invitrogen
(Carlsbad, CA) was performed according to standard PCR conditions and a ABI7900HT
(Applied Biosystems, Foster City, CA) real-time PCR system. Primers for quantitative
RT-PCR are listed in Appendix A (Table A1).
28
3.2.5. Western blotting
Cells were lysed in CelLytic buffer (Sigma, St. Louis, MO) supplemented with 1%
protease inhibitor cocktail (Pierce, Rockford, IL), 1% phosphatase inhibitor cocktail
(Calbiochem) and 1mM PMSF. 10 μg of protein mixture per sample were separated on
NuPAGE (Invitrogen, Carlsbad, CA) 4-12% Bis-Tris gradient gels and transferred on
PVDF membranes (Immobilion, Millipore, Temecula, CA). For the detection of mouse
and human proteins by Western blot, primary antibodies were used together with the
WesternBreeze immunodetection system (Invitrogen). The following antibodies were
used: AID (L7E7, Cell Signaling Technology), FoxO4 (9472, Cell Signaling Technology),
RAG1 (a kind gift from Dr. David Schatz’s Laboratory), STAT5 (3H7, Cell Signaling
Technology) and phospho-Y694 STAT5 (14H2, Cell Signaling technology). Antibody
against -Actin was used as loading control (Santa Cruz Biotechnology– C4-SC47778).
3.2.6. Flow cytometry and Cell cycle analysis
Antibodies against mouse B220 (RA3-6B2), c-Kit (2B8), Sca-1 (D7), CD19 (1D3), CD25
(7D4), CD43 (S7), IgM or µ-chain (R6-60.2 and II/41) and κ light chains (187.1) as well
as respective isotype controls were purchased from BD Biosciences, San Jose,
California. Anti-mouse IL-7Rα (A7R34) was purchased from eBioscience (San Diego,
CA).
3.2.7. Exome Capture Array to detect mutations
Mutation analysis of murine leukemia samples were carried out using a mouse exome
capture array (Sure SelectXT Mouse All Exon Kit, Agilent Technologies) which covers
221,784 exons spanning 24,306 genes.
Bone marrows isolated from leukemic and non-leukemic mice were used for the exome
capture. Genomic DNA was isolated from each sample, sonicated and hybridized to the
prepared exon library. Sequencing was carried out using Illumina Technology. The
29
sequences were then mapped back to the mouse genome and common SNPs
eliminated from the analysis using dbSNP.

3.3. Results
3.3.1. Link between AID and pre-B ALL
A meta-analysis of genes which are frequent hypermutation targets of AID in mature
germinal center B cells revealed that they are also commonly mutated, deleted or
amplified in childhood ALL (Figure 3.1). This implied a possible role for AID in pre-B ALL
progression.
Figure 3.1. Correlation between hypermutation targets of AID and childhood ALL.
On plotting the genes which are favorite hypermutation targets of AID in Ung
-/-
Msh2
-/-

splenic B cells induced with LPS and IL4 (Liu et al., 2008), against genes which are
commonly deleted and amplified in childhood ALL (Mullighan et al., 2007), a positive
correlation was observed.

3.3.2. Early murine B cells are safeguarded from pre-mature AID activation
Little is known about AID expression at early stages of B cell development. We therefore,
wanted to identify the subset of early B cells that express AID at biologically significant
30
levels. Following this, we investigated the mechanism by which AID expression was
prevented before this stage of B cell development and the signal transduction pathways
and transcription factors involved in this process of safeguard. To this end, we sorted
bone marrow extracted from a wildtype mouse into various fractions of B cell
development based on Hardy’s Principle (Figure 3.2A) and measured mRNA levels of
AID in each fraction by qRT-PCR (Figure 3.2B). We observed that AID mRNA levels
increase significantly when cells differentiate from Fraction C’ (large pre-B) to Fraction D
(small pre-B) stage of development.
Figure 3.2. Early B cells are safeguarded from pre-mature AID activation until
Fraction D.
A: Sorting of wildtype murine bone marrow into different fractions of early B cell
development based on Hardy’s Principle. B: Quantitative RT-PCR showing Aid mRNA
levels in each fraction of early B cell development, as compared to splenic B cells
induced with LPS and IL4 (positive control) and Aid
-/-
pre-B cells (negative control).

3.3.3. Identification of IL7R signaling as the molecular mechanism safeguarding murine
pre-B cells from pre-mature AID activation
When B cells differentiate from Fraction C’ to D they downregulate surface IL7R as a
consequence of the signals emanating from the pre-B cell receptor (Figure 3.3A). We
thus tested whether Aid mRNA levels are increased at Fraction D by differentiating
31
Fraction C’ pre-B cells in vitro using 2 methods: 1) By retroviral reconstitution of
functional (SLP65/GFP) and non-functional (SLP65
Y96F
/GFP) versions of SLP65 into
Slp65
-/-
pre-B cells, and measuring Aid mRNA and protein levels (Figure. 3.3B-C) and 2)
by direct IL7 withdrawal from pre-B cell cultures and measuring Aid mRNA and protein
levels (Figure. 3.4A-B).
In vitro differentiation of pre-B cells by both methods led to a significant upregulation of
AID, showing that IL7R signaling safeguards early B cells before Fraction D from AID
expression.
Figure 3.3. Pre-B cell receptor signaling upregulates AID at Fraction D by
downregulating surface expression of IL7Rα.
A: FACS analysis of surface IL7Rα in Slp65
-/-
pre-B cells expressing empty GFP vector,
SLP65/GFP or SLP65
Y96F
/ GFP. B: Quantitative RT-PCR showing Aid mRNA levels
upon reconstitution of Slp65 into Slp65
-/-
pre-B cells. C: Western blot depicting increase
in AID protein level upon differentiation of pre-B cells from Fraction C’ to D.

32
Figure 3.4. In vitro differentiation of early B cells from Fraction C’ to D by IL7
withdrawal upregulates AID.
A. Aid mRNA levels measured by qRT- PCR before and after 24 hours of IL7 withdrawal.
B. AID protein levels measured by western blotting before and after IL7 withdrawal.

3.3.4. IL7R signaling safeguards human pre-B cells from pre-mature AID expression
In order to test the hypothesis that IL7R signaling plays a protective role against AID
expression in human pre-B cells, we carried out a comprehensive analysis of bone
marrow samples from two children who had mutations in either one of the chains of the
receptor, namely the I L 7R α chain or the common I L2 R γ chain. The bone marrow
samples were sorted into Fractions C’ and D (Figure 3.5A) followed by quantification of
AID mRNA levels and measurement of mutation frequency of V
H
gene in each fraction
(Figure 3.5B). V
H
is the most frequently hypermutated target of AID, thus enabling us to
measure the functionality of AID in a setting where IL7R signaling is compromised.
Interestingly, we observed that IL7R signaling also protected human pre-B cells from
pre-mature AID expression, thus making it important in a setting where childhood
leukemia may develop.

33
Figure 3.5. IL7R signaling safeguards human pre-B cells from premature AID
expression before Fraction D.
A: Representative FACS plots showing sorting of bone marrow from children with
mutations in IL7R into Fractions C’ and D. B: Plots showing AID mRNA levels (left) and
V
H
mutation frequency (right) in the two fractions, in each child with the mutation in one
of the chains of the IL7R. Normal human bone marrows were used as negative controls.
(Experiment perfomed in collaboration with Chaim Roifman, Rebecca Buckley and Klaus
Schwarz).

3.3.5. Identification of JAK-STAT and PI3K signaling pathways as negative regulators of
pre-mature AID expression
Following our identification of IL7R as the key signaling molecule blocking the pre-
mature AID expression in pre-B cells, we carried out a comprehensive study of the
transcription factors which are negatively and positively regulating AID expression before
and at Fraction D respectively.
34
As STAT5 is directly downstream of IL7R signaling (Goetz et al., 2004), we assessed
the effect of ablation of Stat5 on AID expression levels. After confirming that STAT5 was
the putative negative regulator of AID expression (Figure 3.6), we next investigated
which transcription factor activates AID at Fraction D.
Figure 3.6. STAT5 is a negative regulator of AID expression.
IL7-dependent pre-B cells were isolated from a Stat5
fl/fl
mouse and conditional deletion of
Stat5 was carried out using a retroviral vector that inducibly expresses Cre-recombinase.
AID protein levels in Stat5
fl/fl
IL7-dependent cells transduced with empty ER
T2
vector and
Cre-ER
T2
vector were compared by western blotting.

Since our preliminary studies showed that SLP65 activates AID downstream of pre-BCR
(Figure 3.3B-C), we hypothesized that FoxO transcription factors which are downstream
of SLP65, bind to Aid promoter and turn on its transcription. Interestingly, active nuclear
FoxO4 was upregulated when pre-B cells were differentiated from Fraction C’ to D
(Figure 3.7B). In addition, our experiments showed that constitutively active forms of
FoxO1 and FoxO3a turn on Aid expression in Fraction D pre-B cells (Figure 3.7 C-D).
Finally, the common upstream activator of all the FoxOs, i.e. PTEN, was evaluated to
check if it played a role in the activation of Aid expression at Fraction D. We observed
that loss of PTEN by inducible deletion abrogated expression of Aid mRNA in pre-B cells
(Figure 3.7A).
35
Figure 3.7. FoxO factors and PTEN are transcriptional activators of Aid at Fraction
D.
A: Loss of PTEN abrogates Aid expression in murine pre-B cells. IL7-dependent pre-B
cells derived from the bone marrow of a conditional knockout mouse for Pten were
retrovirally transduced with an empty puromycin vector and Cre-ER
T2
puromycin vector.
Deletion of PTEN was induced by tamoxifen which activated the Cre recombinase in the
Pten
fl/fl
pre-B cells transduced with Cre-ER
T2
vector. Aid mRNA levels were then
compared between the empty vector transduced cells and PTEN deleted cells 48 hours
after tamoxifen induction, by qRT-PCR. B: Slp65
-/-
IL7-dependent pre-B cells were
differentiated from Fraction C’ to D by Slp65 reconstitution and protein level of nuclear
FoxO4 was measured. C: Murine IL7-dependent pre-B cells were retrovirally transduced
with either an empty vector or a constitutively active form of FoxO1 (FoxO1
CA
). The 2
groups of cells were then subject to two conditions each, either they were retained in the
presence of IL7 (Fraction C’) or IL7 withdrawal was carried out for 24 hours to
differentiate them to Fraction D. Aid mRNA level was then measured in each case by
qRT-PCR. D: An identical experiment to one described in 3.7C was performed using a
retroviral vector for constitutively active FoxO3a (FoxO3a
CA
).

36
3.3.6. Fraction D pre-B cells respond to inflammatory signals

AID is generally upregulated in a mature B cell upon antigen encounter (Muramatsu et
al., 2000). We thus tested whether Fraction D pre-B cells respond to an inflammatory
agent like LPS, by upregulating AID in a manner similar to mature B cells. To this end,
we used pre-B cells from AID-GFP reporter mouse strain in which a portion of the fifth
exon of AID has been replaced with GFP (Figure 3.8A; Crouch et al., 2007). AID-GFP
pre-B cells were maintained in the presence of IL7 or differentiated by IL7 withdrawal,
and each case was either treated with LPS or left unexposed to the inflammation-
inducing agent. Time-dependent flow cytometry was then used to analyze the change in
percentage of GFP
+
cells in each case, which is a direct readout for the amount of AID
present (Figure 3.8B). AID-GFP pre-B cells which were differentiated to Fraction D by
IL7 withdrawal showed the maximum expression of AID upon LPS treatment, indicating
that Fraction D pre-B cells respond to infectious/ inflammatory cues (here, LPS
treatment; Figure 3.8B-C). Staining for surface κ light chain was carried out to ensure
that a portion the AID-GFP
+
cells were small pre-B cells and had not differentiated all the
way to form immature cells. We observed that a large fraction of the AID-GFP
+
cells
were negative for surface κ expression, thus confirming that AID is first upregulated
when pre-B cells differentiate from Fraction C’ to D (Figure 3.8C).
An identical experiment to the one described above was also performed with AID Cre-
YFP IL7-dependent pre-B cells (Figure 3.9A; Crouch et al., 2007), which showed that
Fraction D pre-B cells responded to LPS and showed the highest percentage of YFP
+

cells (Figure 3.8B-C). In AID Cre-YFP mice, Cre recombinase is placed under the control
of the AID promoter. Activation of AID turns on Cre expression which excises out the Lox
Stop Lox (LSL) cassette preceding YFP at the ROSA26 locus, thus allowing YFP to be
expressed (Figure 3.9A). Unlike the AID-GFP cells described above, AID Cre-YFP cells
37
do not capture the expression of AID in real time. These cells would get permanently
labeled with YFP once AID has been expressed.
Figure 3.8. Fraction D cells from AID-GFP reporter mice respond to a surrogate of
infection like LPS by upregulating AID.
A: A diagrammatic representation of the construct in AID-GFP reporter mouse strain. B:
Change in percentage of AID-GFP
+
cells with time measured by flow cytometry, before
and after differentiation to Fraction D, both in the presence and absence of LPS
treatment. C: Representative FACS plots depicting the percentage of AID-GFP
+
cells in
Fraction D upon LPS treatment.

38
Figure 3.9. Fraction D cells from AID Cre-YFP reporter mice respond to
inflammatory signals from LPS by upregulating AID.
A: A diagrammatic representation of the construct in AID Cre-YFP reporter strain. B:
Change in percentage of AID Cre-YFP
+
cells with time, in the presence and absence of
LPS, before and after differentiation to Fraction D. C: Representative flow cytometry
plots showing percentage of cells that responded to LPS treatment (YFP
+
) before and
after differentiation to Fraction D.

We also measured the mRNA and protein levels of AID at Fractions C’ and D in
response to inflammatory signals like LPS (Figure A.3) and obtained similar results to
the ones described above in Figures 3.8 and 3.9.
One of the key negative regulators of AID abundance and expression in the germinal
center is miR-155 (Teng et al., 2008). We therefore examined the levels of miR155 in
Fraction C’ and Fraction D pre-B cells in response to inflammation (here, LPS).
Interestingly, unlike AID, miR-155 levels do not increase when cells differentiate from
Fraction C’ to D (Figure 3.10). In addition, we observed that AID upregulation at Fraction
39
D was not accompanied by miR-155 upregulation even upon LPS treatment (Figure
3.10). Hence, we conclude that pre-mature AID expression during early B cell
development cannot be countered by mechanisms that commonly downregulate its
expression in the germinal center. Therefore, AID expression upon inflammation at
Fraction D would go unchecked and allow the accumulation of deleterious mutations in a
Fraction D pre-B cell leading to full-blown leukemia.
Figure 3.10. Fraction D cells lack protective mechanisms against pre-mature AID
activation.
Levels of miR-155 were measured using qRT-PCR and normalized to the level of the
control miRNA (here, snoRNA202). miRNA isolation and PCR were carried out
according to the miRvana protocol provided by Invitrogen.

3.3.7. Fraction D – The subset most vulnerable to leukemogenesis
We and others have previously published that IL7 withdrawal in pre-B cells results in
upregulation of Rag1 and Rag2 which are required to rearrange the light chain at
Fraction D (Figure 3.11A; Duy et al., 2010). Rag1 and Rag2 are upregulated by the
PTEN-FoxO pathway at Fraction D (Figure 3.11 B-E). It is particularly interesting to note
that both AID and the RAG recombinases peak during early B cell development at
Fraction D, and are regulated by the same signaling molecules. A flow chart illustrating
the signaling pathways that regulate Aid and Rag1/ Rag2 at the transition from Fraction
40
C’ to D is shown in Appendix A (Figure A.2). Such concomitant expression of AID and
RAG enzymes in the same cell (Fraction D), pointed to a cooperative mechanism
leading to leukemia progression.
Our results collectively indicate that Fraction D would be the subset of increased genetic
vulnerability for two reasons- 1) It lacks safeguards like IL7R and miR155 that prevent
accumulation of AID-induced mutations and, 2) It displays concomitant expression of
AID and RAG enzymes.
Figure 3.11. Rag1/Rag2 recombinases are upregulated by the PTEN/FoxO pathway
when cells differentiate from Fraction C’ to D.
A: Western blot showing RAG1 and AID upregulation after 3 days of IL7 withdrawal from
Fraction C’ pre-B cells. B: Loss of PTEN by inducible deletion prevents upregulation of
Rag1 mRNA. C: Loss of PTEN by inducible deletion prevents upregulation of Rag2
mRNA. D: Upregulation of Rag1 mRNA upon retroviral expression of a constitutively
active form of FoxO1 (FoxO1
CA
) in Fraction C’ pre-B cells. E: Upregulation of Rag2
mRNA upon expression of a constitutively active form of FoxO1 (FoxO1
CA
) in Fraction C’
pre-B cells.

41
3.3.8. Pre-mature AID activation promotes the clonal evolution of a B cell clone in the
bone marrow of a child carrying the TEL-AML1 rearrangement.

TEL-AML1 gene rearrangement arises in utero but only <1% of the children who carry
the rearrangement develop full-blown leukemia (chapter 2). This would imply that TEL-
AML1 driven leukemia would require a second hit. We hypothesized that AID causes the
second hit leading to leukemic transformation of a murine pre-B cell carrying the TEL-
AML1 rearrangement.
In order to test the above hypothesis directly, we carried out an in vivo leukemia
transplantation experiment in NOD SCID γ
C
mice, with TEL-AML1 Tg pre-B cells (Ford et
al., 2009) expressing either control vector or tamoxifen-inducible AID overexpression
vector. All mice injected intrafemurally with the TEL-AML1 Tg AID-ER/GFP pre-B cells
developed full blown leukemia, while those injected with TEL-AML1 Tg GFP pre-B cells
lived on, confirming the role of AID in accelerating a TEL-AML1-driven leukemia (Figure
3.12).
Figure 3.12. AID overexpression accelerates TEL-AML1-driven leukemia in vivo.
A: PCR was carried out on cDNA isolated from wildtype (WT) control and TEL-AML1
transgenic (Tg) mice to verify the expression of TEL-AML1 in the latter. REH, a human
cell line known to carry the TEL-AML1 rearrangement was used as the positive control.
B: Luciferase bioimaging of NOD-SCID γ
C
mice injected with either TEL-AML1 Tg IL7-
dependent pre-B cells containing empty vector or with TEL-AML1 Tg AID ER/ GFP-
containing IL7-dependent cells. C: Kaplan Meier curves comparing the overall survival
percentage of mice in both groups.

42
Figure 3.12. continued.

3.3.9. Genetic ablation of Aid and Rag1 abrogates leukemia initiation upon inflammation
To test the hypothesis that AID and RAGs are required for the leukemic transformation
of TEL-AML1 carrying pre-B cell clones in the context of inflammation, we subjected
TEL-AML1 expressing pre-B cells from wildtype mice, Aid
-/-
, and Rag1
-/-
mice to five
rounds of IL7 withdrawal and LPS treatment. These cells were then labeled with firefly
luciferase and injected intravenously into NOD SCID mice. All the mice which received
wildtype pre-B cells with IL7 withdrawal and LPS treatment died of leukemia within 18
days. On the contrary, genetic ablation of either Aid or Rag1 delayed or abrogated
leukemia development even after 5 cycles of IL7 withdrawal and LPS treatment (Figure
3.13 A-D). The above described experiment highlights the requirement of both AID and
RAGs for the leukemic transformation of a Fraction D pre-B cell. In addition, it
corroborates our previous findings which identify IL7R as the guardian of the B cell
genome till Fraction D.
43
Figure 3.13. AID and RAG1 are required for the leukemic transformation of TEL-
AML1 carrying pre-B cell clones in the context of inflammation.
A: Luciferase bioimaging of NOD-SCID mice injected with Aid
+/+
Rag1
+/+
+IL7, Aid
+/+

Rag1
+/+
-IL7+LPS, Aid
-/-
+IL7, Aid
-/-
-IL7+LPS, Rag1
-/-
+IL7 and Rag1
-/-
-IL7+LPS, with all
cell types overexpressing TEL-AML1/GFP. B: Kaplan Meier curves comparing the
overall survival percentage of mice in all 6 groups. C: Spleen and liver were isolated
from mice in the Aid
+/+
Rag1
+/+
-IL7+LPS group and subjected to H and E staining and
CD19 immunohistochemistry to verify that the cause of death was pre-B ALL. D:
Verification of leukemia was also carried out by flow cytometry measurement of the
percentage of CD19
+
/ TEL-AML1- GFP
+
cells in the bone marrow and spleen of all the
sacrificed mice in the Aid
+/+
Rag1
+/+
-IL7+LPS group.

44
3.3.10. Repeated exposure of Fraction D cells to inflammation results in genetic
instability

Our studies both in vitro and in vivo confirmed AID and RAG1/RAG2 as the missing links
between inflammation and leukemic transformation of a TEL-AML1
+
pre-B cell. Based on
our previous results, we concluded that AID and the RAG enzymes were required for the
leukemic transformation of TEL-AML1 carrying pre-B cell clones in the context of
infection. The next approaches were aimed at characterizing the changes that occur at
genomic level as a result of pre-mature AID activation in pre-B cells, upon inflammation
or infection.
Our first approach involved spectral karyotyping (SKY) analysis of the bone marrows of
two leukemic mice from the Aid
+/+
Rag1
+/+
-IL7+LPS group. Through this, we wanted to
identify and characterize the chromosomal abnormalities (aneuploidy and translocations)
present in the cell. We observed that nearly all analyzed bone marrow cells from both
leukemic mice shared one abnormality, namely the trisomy of chromosome 18 (Figure
3.14; Figure A.4). In addition, each leukemia also contained additional secondary
genetic changes like translocations, deletions etc. Based on these results, we drew a
genealogy tree depicting the clonal architecture of the leukemia that originated from
TEL-AML1 containing pre-B cells exposed to repeated inflammatory cues at Fraction D
(Figure 3.14).

45
Figure 3.14. mSKY reveals clonal architecture of TEL-AML1 pre-B ALL.
SKY analysis was performed on the bone marrows of two leukemic mice in the Aid
+/+

Rag1
+/+
-IL7+LPS group, according to standard protocol (Padilla-Nash et al., 2007). 18
single cells per mouse were analyzed to identify the chromosomal aberrations that each
carried. Based on the analysis, a genealogy tree was drawn to check if the leukemia was
clonal in origin.

The study that followed this was aimed at identifying the exact mutations that had
occurred in the leukemic clones upon IL7 withdrawal and LPS treatment in the group
that received the Aid
+/+
Rag1
+/+
cells. Two of the mice which received Aid
-/-
-IL7+LPS
cells also developed leukemia, though much later than the wildtype group. Therefore,
one representative mouse from this group was included as control in the mutation
analysis to subtract any AID-independent mutation that may have been acquired in the
wildtype sample upon inflammation. Similarly, one bone marrow sample from the Rag1
-/-

-IL7+LPS group was included to subtract any RAG1-independent genomic changes. We
carried out this mutation analysis using a mouse exome capture array (described in
Materials and Methods, Chapter 3).
46
Bone marrow isolated from one mouse in each group (Aid
+/+
Rag1
+/+
-IL7+LPS, Aid
-/-
-
IL7+LPS and Rag1
-/-
-IL7+LPS) were used for the exome capture. After the elimination
of common SNPs, genetic alterations were compared between the samples to eliminate
those which were common to both Aid
+/+
and Aid
-/-
condition, thus getting rid of the AID-
independent mutations. The SNPs that remained after these rounds of processing were
analyzed to identify a pattern 1) either as being common AID hypermuation targets (Liu
et al., 2008), or 2) involved in early B cell development and leukemia.
Preliminary analyses revealed the presence of point mutations in genes frequently
altered in childhood pre-B ALL and also in genes which are frequent hypermutation
targets of AID in the Aid
+/+
Rag1
+/+
-IL7+LPS group (Table 3.1). Analysis to identify
mutations in the Rag1
-/-
-IL7+LPS group are currently ongoing.
Table 3.1. AID-dependent mutations in TEL-AML1 pre-B ALL.
A list of mutated genes in the Aid
+/+
Rag1
+/+
-IL7+LPS group, which are either direct
hypermutation targets of AID or are involved in pre-B ALL are highlighted in red.

47
3.4. Conclusions
As discussed in Chapter 1, checkpoint controls are crucial during B cell differentiation to
thwart leukemogenesis. In Chapters 2 and 3 of this thesis, we elucidate the protective
function of IL7R signaling at the pre-BCR checkpoint (Fraction C’). We describe how it
safeguards the B cell genome from pre-mature acquisition of genetic alterations, thereby
protecting it from leukemogenesis. We prove that IL7R signaling protects pre-B cells
until Fraction D from acquisition of mutations, by blocking the expression of two
enzymes that contribute to genetic instability, namely AID and RAG1/RAG2. We also
show that this protective role of IL7R is also relevant to human pre-B cells.
We then discuss in the detail the molecular players and their interactions which lead to
AID expression at Fraction D. We claim that these AID levels are insufficient to allow the
acquisition of mutations, and therefore, propose that exposure to inflammation or
infection at this stage would increase AID to deleterious levels that are sufficient to
induce leukemogenesis. We have carried out all our studies using LPS
(lipopolysaccharide), a component of gram-negative bacteria to mimic inflammation/
infection. Our experiments demonstrate that inflammation is a major causative agent for
leukemogenesis especially in a scenario where pre-B cells already contain a genetic
alteration which they acquired in utero (eg. TEL-AML1). We prove experimentally the
‘Delayed Infections Hypothesis’ proposed by Dr. Mel Greaves and Dr. Joseph Wiemels,
which states that repeated severe infections pre-dispose a child with a TEL-AML1
rearrangement to leukemia.
In addition, we discuss the importance of the RAG enzymes in the process of acquisition
of chromosomal aberrations. Tsai et al. propose cooperation between AID and RAGs for
leukemogenesis to occur. However, they do not provide information on the stage in early
B cell development or the molecular processes that enable concomitant AID and RAG
expression. In our study, we demonstrate that regulation and pattern of expression of the
48
RAG1/ RAG2 enzymes is highly similar to AID at Fraction D. This would mean that
Fraction D is the subset of early B cells with increased genetic vulnerability. Further
experiments are underway to study the cooperation between AID and RAG enzymes in
the acquisition of genetic alterations.
We conclude that AID and RAGs cooperate to generate the second hit in leukemia,
especially in those sub-groups that rely on additional genetic changes for transformation
(Figure A.1). More importantly, our findings provide a novel insight into signaling
mechanisms at the pre-BCR checkpoint that protect against genetic instability and hence,
leukemogenesis (Figure A.1).

3.5. Limitations and future perspectives
The studies described in Chapter 3 show that Fraction D pre-B cells represent the
subset of increased genetic vulnerability due to rise in AID and RAG1/ RAG2 expression
levels. However, studies proving the expression of AID and RAG1/ RAG2 enzymes in
the same Fraction D cell are lacking. In order to bridge this gap, experiments which allow
the measurement of both AID and RAG activity in a single Fraction D pre-B cell must be
conducted, to prove that both enzymes cooperate in causing overt leukemic
transformation. Such studies are currently ongoing.
Our study describes only a small subset of transcription factors which regulate AID
expression at the Fraction C’ to D transition upon loss of IL7R signaling. However, it
must be noted that AID and Rag1/2 expression is not solely regulated by these factors
(here, Stat5 and FoxO). The role of a number of previously described transcription
factors like PAX5 in regulating AID expression (Xu et al., 2007) has not been elucidated
at Fraction D, and would therefore be interesting for future studies.
Another pitfall in this study is the use of AID overexpression system to assess the
acceleration of TEL-AML1-driven leukemogenesis. Such an overexpression system
49
does not reflect the actual AID levels that are present in a small pre-B cell upon
inflammation. Moreover, overexpression of AID in the absence of TEL-AML1 could lead
to leukemia in mice. In order to rule out this possibility, we need to use a system where
AID overexpression is carried out in the presence of non-functional TEL-AML1 which
lacks the RHD DNA binding domain. A comparison of latency in leukemia initiation in the
two groups would highlight the importance of TEL-AML1 as the intiating hit and AID as
the accelerating factor in leukemogenesis.
In this study, we use LPS as a surrogate for infection and as a trigger for inflammation to
assess pre-mature AID expression in Fraction D cells. In addition, it would be interesting
to test the effect of other inflammatory or infectious agents which are more relevant to
humans. An example would be the antigenic components of the influenza virus which
affect a number of young children. Such studies would provide concrete evidence for the
Delayed Infections Hypothesis.
Finally, assessment of large panels of childhood ALL patients for combined AID and
RAG1/RAG2 mutation signatures would be required to prove that the study described in
this thesis is relevant to the human leukemogenesis.

50
Chapter 4. Diverse roles of BACH2 in B-lymphoid cells- Relevance to B-ALL

In the previous chapters, we highlighted the molecular players at the pre-BCR
checkpoint that guard an early B cell from leukemic transformation by preventing genetic
instability. In the current chapter and in the ones that follow, we describe an alternative
mechanism of leukemogenesis which stems from the loss of apoptotic controls in a
deleterious pre-B cell.
Checkpoints during early B cell development ensure that B cells which do not meet the
requirements at every stage are eliminated by apoptosis and removed from the
repertoire (chapter 1). PAX5 mediates the first checkpoint in this process of quality
control. One of the key molecules transcriptionally activated by PAX5 is the B-lymphoid
transcription factor BACH2 (Casolari et al., 2012). The role of BACH2 in mature B cells
has been extensively studied and is discussed in the following sections of this chapter.
However, BACH2’s role in early B cell development and pre-B leukemogenesis is still
unclear.
As PAX5 is an absolute requirement for B cell development and checkpoint control
(chapter 1), we hypothesized that BACH2 being downstream of this, would also be
involved in quality control of early B cells. Therefore, using BACH2 as an example, we
drive home the importance of negative selection (apoptosis) at the pre-BCR checkpoint
in preventing leukemia initiation (chapter 5; chapter 6).

4.1. Location and structure of BACH2
BACH2 is BTB (Bric à Brac Tramtrack and Broad Domain Complex) And CNC (Cap N
Collar) Homology 1 Basic Leucine Zipper (bZip) Transcription Factor 2 (Oyake et al.,
1996). It is a transcription factor that belongs to the basic leucine zipper family.
51
BACH2 maps to chromosome 6q15. It has a telomere to centromere transcriptional
orientation. Human BACH2 has 9 exons with the first five being non-coding and exons 6
to 9 coding for the full length protein. Mouse BACH2 protein is 97.5% homologous to
human BACH2 and has 7 exons. Exons 4 to 7 are the coding exons of mouse Bach2
gene.
As the name suggests, BACH2, possesses 2 important domains: BTB and bZip. There is
also a serine rich domain which precedes the bZip domain. The BTB, bZip and the
serine rich domains are highly conserved in different species. Each protein motif in
BACH2 has a unique function. The BTB domain, located at the amino terminus is
required for protein-protein interaction (Oyake et al., 1996). It is also involved in oligomer
formation, DNA loop generation and nuclear foci generation. By virtue of its bZip domain
which is rich in basic amino acids like arginine and lysine, BACH2 mediates direct DNA
binding (Muto et al., 2002). BZip domain of BACH2 also possesses the nuclear
localization signal (NLS) (Hoshino et al., 2000) which is essential for BACH2 to function
as a transcription factor.

4.2. BACH2 is a B-lymphoid transcription factor
In order to study if BACH2 expression varied with different stages of B cell development,
Muto et al. measured BACH2 protein levels at every stage. Within the hematopoetic
lineages, BACH2 was expressed only in B-lymphocytes. A comparison of stage specific
expression of BACH2 during B cell development revealed that it was present at every
stage except in plasma cells. Its expression was high during pre-pro-B stage of
development. Studies with plasmocytoma cell lines revealed a loss of BACH2
expression corroborating the findings made in plasma cells (Muto et al., 1998).
Further studies were conducted to understand the reason and the mechanism behind
the loss of BACH2 expression at the plasma cell stage. This led to the identification of
52
the crucial role played by BACH2 in germinal center formation after antigen encounter.
Experiments carried out using a knockout mouse model for Bach2 revealed the
importance of BACH2 in delaying the process of plasma cell differentiation by
transcriptional repression of plasma cell factor Blimp1. PAX5 which is the master
regulator of B cell differentiation (chapter 1) activates Bach2 and Aid, following which,
BACH2 represses Blimp1 and increases the time span for somatic hypermutation and
class switch recombination by AID to occur (Muto et al., 2010).
Based on these observations, it was not a surprise that Bach2
-/-
mice had defective
germinal center formation. Complementary determining regions (CDR) regions in the
knockout mice revealed absence of somatic hypermutations and the mice failed to class
switch and produce IgG upon antigen encounter. Bach2
-/-
mature B cells have lower AID
and PAX5 levels and increased amount of plasma cell factors like BLIMP1 and XBP1
(Muto et al., 2004).

4.3. BACH2 can act as a transcriptional repressor and activator
BACH2 binds to the TPA (12 O-Tetra decanoylphorbol-13-acetate) response element
(TRE) on the DNA. Bach2 was identified from a Y2H screen for MafK (Oyake et al.,
1996). In addition to TRE, BACH2 also binds to the related MARE (MAF Response
Element) and ARE (Antioxidant Response Element) as homodimers or in combination
with other MAF proteins. TRE, MARE and ARE elements share the same consensus
sequence (TGAG/CTCA) and bind proteins belonging to the MAF family (Muto et al.,
2002).
Activation or repression of genes at the above elements is governed by 2 factors,
namely, the member of the MAF family and its protein binding partner that bind to the
consensus sequence. Transcriptional activation of genes downstream of all three DNA
elements occurs by binding of MAF factors in combination with CNC proteins like NRF1
53
and NRF2. As a result, key genes involved in evoking an antioxidant response to cellular
stress are upregulated, for example, glutathione S-transferase (GST), peroxiredoxin and
hemoxygenase1. Transcriptional repression at these loci is initiated by binding of
heterodimers of BACH1 or BACH2 with one of the MAF proteins (Muto et al., 2002). This
triggers a program of oxidative stress-induced cell death which will be discussed in the
following section.
MAF proteins are divided into big MAFs and small MAFs. The big MAFs, namely, cMAF,
MAFB and NRL contain transactivation domains in addition to bZip domains and thereby,
bind to MARE elements directly, and activate transcription at these regions. In contrast,
the small MAFs, namely, MAFG, MAFK and MAFF possess only a bZip domain and no
transactivation domain. Therefore, the small MAFs form heterodimers with proteins like
BACH1 and BACH2 and repress transcription (Sasaki et al., 2000).
Although widely characterized as a repressor, BACH2 can activate transcription at select
loci. One such situation occurs when BACH2 binds to MAZR, a transcription factor that
contains BTB and zinc finger (ZNF) domains. It has been shown that BTB domain of
MAZR interacts with the BTB domain of BACH2, causing DNA looping and
transcriptional activation of c-Myc. MAZR binds to G rich regions on the c-Myc promoter
followed by binding of BACH2 and MAFK to adjacent MARE elements, which results in
transcriptional activation (Kobayashi et al., 2000).

4.4. BACH2 triggers oxidative stress-induced cell death
As mentioned in Section 4.3, BACH2 is an inducer of oxidative stress-induced cell death.
Extensive studies have been carried out to understand which of these protein motifs in
BACH2 are crucial to mediate its apoptotic function. It has been proved experimentally
that the bZip domain is indispensable for apoptosis. In contrast, removal of the BTB
54
domain does not completely abolish apoptosis, but just reduces the amount of apoptosis
in comparison to wildtype BACH2 (Muto et al., 2002).
BACH2 causes increased apoptosis in the presence of oxidative stressors like DEM and
H
2
O
2
. Most of the BACH2 protein is generally present in the cytoplasm in the absence of
oxidative stress. The extreme C terminal end of the protein contains the cytoplasmic
localization signal (CLS) which masks the nuclear localization signal (NLS) located in the
bZip domain. Cytoplasmic localization of BACH2 is carried out by CRM1/EXPORTIN1. In
the absence of oxidative stress, CLS is more active than NLS and most of BACH2
protein is cytoplasmic. Oxidative stress triggers nuclear localization of BACH2 by
activating NLS and by blocking CRM1/EXPORTIN 1 mediated cytoplasmic localization
(Hoshino et al., 2000).
Any antioxidant response in the cell prevents the nuclear accumulation of BACH2. For
example, the enzyme catalase converts H
2
O
2
to less toxic compounds and prevents the
accumulation of reactive oxygen species (ROS). It has been experimentally shown that
addition of catalase to BACH2 overexpressing cells in the presence of oxidative stressor
drugs prevents nuclear accumulation of the latter and subsequent apoptosis (Kamio et
al., 2003). Another antioxidant protein HEME can bind to BACH2 and inhibit BACH2-
dependent nuclear repression. It does so by interacting with 5 cysteine- proline motifs on
the BACH2 protein, which prevent BACH2 from binding to MARE. HEME also induces
degradation of BACH2 (Watanabe-Matsui et al., 2011).

4.5. Bach2 is a common integration site for viruses
In a study by Liu et al., Bach2 was identified as a common site for retroviral integration
by murine leukemia virus (MLV) (Figure 4.1). Integrations mainly occurred in Bach2
promoter and within the first three introns. Provirus integration led to lower Bach2 mRNA
levels in all samples compared to normal spleen, with one exception. In the exceptional
55
sample, there was higher amount of Bach2 mRNA as compared to normal spleen.
Further investigations revealed an alternative promoter in Bach2 intron 2 which resulted
in alternative transcripts. This transcript was being produced in smaller amounts in
normal samples and all the other tumor samples tested (Liu et al., 2009).
In addition to the alternative transcript resulting from the promoter in intron 2, alternative
exons were identified in intron 4 which resulted in shorter Bach2 transcripts lacking the
bZip domain. This resulting BACH2 protein showed perinuclear localization and was
proposed to act in a dominant negative fashion because its BTB domain could interact
with multiple proteins. All proviral integrations observed in the above study shared two
common features: 1) integrations were solely in Bach2 intronic regions and, 2) the
insertions were in opposite orientation to Bach2 transcription which resulted in
inactivation of wildtype Bach2 (Liu et al., 2009).
Figure 4.1. Positions of MLV integration within the Bach2 locus.
All the common integration sites (CIS) represented by blue circles are located in Bach2
intronic regions. Exons are depicted using shaded blocks and the interspersing introns
using a straight line. (Diagram not drawn to scale).

In an independent study, MLV was found integrated into the Bach2 locus in antisense
orientation. This led to lower expression of canonical Bach2 and increased expression of
the antisense transcript. Expression of the antisense transcript was driven by the 5’LTR
of the virus which resulted in inhibition of RNA polymerase, thus blocking the expression
of the sense transcript (Rasmussen et al., 2010).
56
A lymphoma line named Raji showed no expression of BACH2. Investigations carried
out to identify the reason behind the absence of BACH2 expression revealed an
integration of Epstein-Barr Virus (EBV) into exon 1 of BACH2. EBV is an episomal virus
during latent infection, and only integrates into the genome during a persistent infection.
In Raji cell line, both the paternal and maternal chromosomes were disrupted by
integration of EBV at 6q15 leading to loss of BACH2 expression. It was observed that
the integration occurred at a GC rich region by homologous recombination. Loss of
BACH2 expression has been proposed to be the reason for increased clonogenecity and
tumorogenecity of Raji. So far, Raji is the only burkitt lymphoma cell line where a loss of
BACH2 expression has been reported (Takakuwa et al., 2004).
Another study aimed at identifying mutations which cooperate with nucleophosmin to
result in AML, showed that Bach2 was frequently inactivated by insertional mutagenesis
(Vassiliou et al., 2011). All these studies highlight the fragility of the Bach2 locus which
makes it a hot target for viral integrations and transpositions.

4.6. BACH2 is a putative tumor suppressor in B cell lymphoma
Deletions of chromosome 6q have been observed in high grade B cell lymphomas. In a
patient-derived B cell lymphoma line BLUE-1, translocation of the oncogene BCL2L1
was observed downstream on BACH2 exon 1. This translocation disrupted BACH2
function by preventing the expression of full length BACH2. Instead, BACH2 promoter
and regulatory elements in BACH2 exon 1 were driving the expression of BCL2L1. It
was hypothesized that the death of the patient from where BLUE-1 line was derived,
occurred due to the loss of BACH2 expression combined with an increased expression
of BCL2L1 (Türkmen et al., 2011).
In a study conducted by the Osaka Lymphoma Study Group in Japan, BACH2 level was
identified as an independent prognostic factor deciding clinical outcome in diffuse large
57
B cell lymphoma (DLBCL) patients. Patients in intermediate and high risk groups were
found to have different overall survival (OS) rates based on their status of BACH2
expression. BACH2 expression level was measured by cytoplasmic staining of
lymphoma samples from the patients. Patients with increased cytoplasmic staining of
BACH2 showed favorable outcome and good response to doxorubicin treatment.
Doxorubicin induces oxidative stress and nuclear accumulation of BACH2, thus leading
to subsequent apoptosis of the lymphoma cells in these patients. This might explain why
patients with higher BACH2 expression responded better to doxorubicin chemotherapy
and survived longer than those with lower BACH2 expression (Sakane-Ishikawa et al.,
2005).
Another study conducted on lymphoma patients with BCL2-IgH fusion highlighted the
importance of BACH2 levels in deciding both clinical outcome and treatment strategy. It
was observed that patients with this translocation presented prolonged survival rates
and better response to chemotherapy, if they had higher BACH2 expression levels. In
this study, BACH2 has been proposed to mediate its tumor suppression through
repression of the IgH enhancers and thereby lowering BCL2 expression levels in
patients (Green et al., 2009). As a summary, all known previous studies on lymphoma
have identified BACH2 as a putative tumor suppressor.

4.7. The BACH2 and BCR-ABL1 connection
Studies with the BCR-ABL1 kinase inhibitor imatinib show that BACH2 is upregulated
upon imatinib treatment in Ph
+
lymphoid cell lines (Casolari et al., 2012; Figure C.1).
Imatinib disrupts association of BACH2 with centromeric heterochromatin and induces
its expression. Such an effect is seen only in lymphoid cell lines showing that this
disruption can occur only in the presence of B cell specific factors (Ono et al., 2007). A
recent study demonstrated that BCR-ABL1 transcriptionally suppresses BACH2
58
expression by mediating PAX5. The paper also identified PAX5 as a direct transactivator
of BACH2 (Casolari et al., 2012).
As mentioned in chapter 1, PAX5 is crucial for maintaining B-cell identity. PAX5
mediates V(D)J recombination (Rolink et al., 2000a) and acts as the first checkpoint
ensuring quality control during early B cell development (chapter 1). As PAX5 is a
transcriptional activator of Bach2 (Casolari et al., 2012), we hypothesized that BACH2
may be crucial for quality control at the subsequent checkpoint, namely the μHC
checkpoint.
To date, there are very few studies on the role of BACH2 in Ph
+
and other subgroups of
pre-B acute lymphoblastic leukemia. Therefore, the main focal points of the second part
of this thesis are two-fold. First we dissect the importance of BACH2 at the pre-BCR
checkpoint during B cell differentiation. Next, using the results from our first study
(Chapter 5) as the framework, we explain how loss of BACH2 expression at the pre-BCR
checkpoint leads to leukemia (chapter 6).

59
Chapter 5. BACH2 is required for negative selection at the pre-BCR
checkpoint

5.1. Introduction
BACH2 is indispensable in mature B cells, where it is required for somatic hypermutation,
class switching and germinal center formation (Muto et al., 2004). Previously published
reports show that BACH2 is a B-lymphoid specific factor, which is expressed at all
stages of B cell development except at the plasma cell stage (Muto et al., 1998). PAX5,
which is upstream of Bach2 keeps the levels of the latter high, thereby preventing the
increase in Blimp1 and plasma cell differentiation (Muto et al., 2010). Relationship
between PAX5, BACH2 and BLIMP1 has been discussed in detail in Chapter 4. Almost
all of these studies have focused on the role of BACH2 in mature B cells and B cell
lymphoma. However, the functions of BACH2 during early B cell development have not
been elucidated.
The upstream activator of BACH2, i.e. PAX5, has been well characterized as a
checkpoint and a tumor suppressor in pre-B ALL (Medvedovic et al., 2011; Mullighan et
al., 2007). Hence, we hypothesized that BACH2 by virtue of its B-lymphoid specificity
and being downstream of PAX5 (Casolari et al., 2012), would play a crucial role in
checkpoint control during B cell differentiation, and hence leukemogenesis.
In the current chapter, we provide concrete experimental evidence for the requirement of
Bach2 in the negative selection (apoptosis) of pre-B cells with non-functional IgH
rearrangements and discuss the mechanism by which this occurs. In addition, we
explain how interplay between BACH2 and the pro-survival molecule BCL6, regulates
balance between negative selection and survival at the pre-BCR checkpoint. Finally, we
discuss how BACH2 regulation of RAG1/ RAG2 activity impacts negative selection and
checkpoint control during B cell development.
60
5.2. Materials and Methods
5.2.1. Extraction of bone marrow cells from mice

Bone marrow cells were extracted from young age-matched Bach2
+/+
and Bach2
−/−
mice
(younger than 6 weeks of age). Bone marrow cells were obtained by flushing cavities of
femur and tibia with PBS. After filtration through a 70 μm filter and depletion of
erythrocytes using a lysis buffer (BD PharmLyse, BD Biosciences), washed cells were
either frozen for storage or subjected to further experiments.
5.2.2. Bach2
+/+
and Bach2
-/-
mice
Bone marrow cells from the above mentioned mice (Muto et al., 2004) were collected
and maintained in presence of 10 ng interleukin-7 (IL7) to generate pre-B cells. In certain
cases, for ease in handling, these pre-B cells were then retrovirally transformed
with BCR –ABL1. All BCR –ABL1-transformed ALL cells derived from bone marrow of
mice were maintained in Iscove’s modified Dulbecco’s medium (IMDM, Invitrogen) with
GlutaMAX containing 20% fetal bovine serum, 100 IUml
−1
penicillin, 100
μgml
−1
streptomycin and 50 µM 2-mercaptoethanol. BCR –ABL1-transformed ALL cells
were propagated only for short periods of time and usually not longer than for 2 months
to avoid acquisition of additional genetic lesions during long-term cell culture.
5.2.3. In vitro differentiation assay using BCR-ABL1 Tyrosine Kinase Inhibitors (TKIs)
Imatinib (TKI) was obtained from Novartis Pharmaceuticals (Basel, Switzerland) or from
LC Laboratories (Woburn, MA). It was dissolved in sterile distilled water and stored
at -20ºC. Addition of imatinib for 3 days to BCR-ABL1- transformed pre-B cells leads to
their differentiation to Fraction D and then to Fraction E. Differentiation is measured by
flow cytometry for surface κ light chain (Duy et al., 2010). This assay is useful to study
differentiation of pre-B cells in vitro.
61
5.2.4. Flow cytometry

Antibodies against mouse B220 (RA3-6B2), κ light chain, IgM (R6-60.2), IgD (11-26c.2a)
as well as their respective isotype controls were purchased from BD Biosciences, San
Jose, California.
5.2.5. Clonality analysis and spectratyping of B cell populations
V
H
-DJ
H
gene rearrangements from B cell populations were amplified using PCR primers
specific for the J558 V
H
region gene together with a primer specific for the Cμ constant
region gene. PCR products were then labeled in a run-off reaction using a FAM-
conjugated Cμ constant region primer along with J558 V
H
primer used in the previous
step. The PCR products from the run-off reaction were subsequently analyzed on a
capillary sequencer (ABI3100; Applied Biosystems) by fragment-length analysis.
Sequences of primers used are given in Table B1.
5.2.6. Retroviral transduction
Transfections of retroviral constructs and their corresponding empty vector controls were
performed using Lipofectamine 2000 (Invitrogen) with Opti-MEM media (Invitrogen).
Retroviral supernatant was produced by cotransfecting 293FT cells with the plasmids
pHIT60 (gag-pol) and pHIT123 (ecotropic env; provided by D.B. Kohn, University of
California, Los Angeles, Los Angeles, CA). Cultivation was performed in high glucose
DMEM (Invitrogen) with GlutaMAX containing 10% fetal bovine serum, 100 IU/ml
penicillin, 100 µg/ml streptomycin, 25 mM Hepes, 1 mM sodium pyruvate, and 0.1 mM of
nonessential amino acids. Regular media were replaced after 16 h by growth media
containing 10 mM sodium butyrate. After 8 h of incubation, the media was changed back
to regular growth media. 24 h later, the virus supernatant were harvested, filtered
through a 0.45-µm filter, and loaded by centrifugation (2,000 g for 90 min at 32°C) two
times on 50 µg/ml RetroNectin-coated non-tissue 6-well plates. 2–3 × 10
6
pre–B cells
62
were transduced per well by centrifugation at 500 g for 30 min and maintained overnight
at 37°C with 5% CO
2
before transferring into culture flasks.
5.2.7. Cloning of MSCV Bach2-ER
T2
IRES GFP vector
BACH2 cDNA was amplified from K562 cell line using primers shown in Table B1. The
cDNA was digested using BamH1 and Xho1 and inserted into the backbone obtained
after digestion of MSCV Cre-ER
T2
IRES puromycin vector with BamH1 and Xho1. The
vector obtained after this ligation was digested using BamH1 and EcoR1 to obtain the
Bach2-ER
T2
fragment. Bach2-ER
T2
was then ligated into a BglII and EcoR1 digested
MSCV IRES GFP vector.
5.2.8. Sequence analysis of V
H
-DJ
H
gene rearrangements
B cells from Bach2
+/+
and Bach2
-/-
bone marrow and spleen were isolated by MACS
(Magnetic-activated cell sorting) using anti-CD19 immunomagnetic beads. MACS was
carried out using the protocol provided by Miltenyi Biotec. RNA was isolated from the
MACS-sorted B cells and converted to cDNA. V
H
-DJ
H
gene rearrangements from B cell
populations were then amplified using primers specific for the J558 V
H
region gene
together with a primer specific for the Cμ constant region gene for 35 cycles. Primer
sequences used are given in Table B1. PCR fragments were cloned using a Topo TA
cloning kit (Invitrogen) and sequenced unidirectionally in 96-well format at Genewiz Inc.
Sequence analysis to determine whether the rearrangement was productive or non-
productive was done using IMGT-V Quest. The link to IMGT-V Quest website is as
follows http://www.imgt.org/IMGT_vquest/share/textes/.
5.2.9. Assay to measure RAG1/RAG2 recombinase activity
Recombination by the RAG enzymes was measured using a retroviral reporter system
called the RSS-GFP (Wossning et al., 2006), where GFP is flanked by the recombination
signal sequences (RSS). In the absence of RAG1/RAG2 activity, GFP is not in the
correct orientation to be expressed. RAG enzyme activity in the cell induces cleavage of
63
RSS sequences that flank GFP, thus returning GFP to the right orientation to be
expressed. For example, in the presence of low levels of RAG activity in BCR-ABL1-
transformed murine wildtype pre-B cells, the percentage of GFP
+
cells transduced and
selected for the vector will be ~3-6%. When BCR-ABL1
+
cells are differentiated using
imatinib for 24 hours, the percentage of GFP
+
cells increases dramatically. This is
because the increased RAG1 and RAG2 expression at this stage causes the GFP to
return to the right orientation by recognizing and cleaving the RSS sequences that flank
it.
5.2.10. Western Blotting
Western blotting was carried out as described in Section 3.2.5 (chapter 3). The following
antibodies were used: Arf (Ab80, rabbit polyclonal, Abcam) and p53 (1C12, mouse
monoclonal, Cell Signaling Technology). Antibody against β-Actin (C4, mouse
monoclonal, Santa Cruz) was used as a loading control.
5.2.11. Growth competition assay
The effect of a gene of interest on the growth and apoptosis of cells was measured using
this assay. Cells were transduced with retroviruses to express the gene of interest in
combination with a redundant fluorescent protein like GFP. The same cell type parallelly
transduced with an empty GFP vector served as the control. Labelled cells were then
tracked in a time-dependent manner by flow cytometry to study if the gene of interest
modified their growth or apoptosis rates.

64
5.3. Results
5.3.1. BACH2 induces negative selection of pre-B cells

We first compared the percentage of mature B cells in bone marrows and spleens from
Bach2
+/+
and Bach2
-/-
mice. It was observed that in the bone marrow of Bach2
-/-
mice,
the pool of K
+
B220
+
and IgM
+
B220
+
recirculating B cells was reduced when compared
to wildtype counterparts (Figure 5.1A). A similar pattern was observed in the spleen for
both the K
+
B220
+
fraction and the IgD
+
B220
+
fraction (Figure 5.1C). K
+
B220
+
cells from
bone marrow and spleen of both the wildtype and knockout mice were subjected to
spectratyping analysis to study their clonality (described in Methods Section 5.2.5).
While κ light chain
+
B cells from Bach2
+/+
bone marrow exhibit a diverse polyclonal
repertoire, κ light chain
+
B cells from Bach2
-/-
mice represent an oligoclonal population
with defective reading frame selection (Figure 5.1B). A similar pattern was observed on
comparing the splenic clonality of the two groups (Figure 5.1D). Normal V(D)J
recombination results in random distribution of the length of V(D)J junctions. Based on
random V(D)J recombination, two of three pre-B cell clones carry out-of-frame
rearrangements, fail to express a functional pre-B cell receptor and are negatively
selected. We observed that effective negative pre-B cell selection results in a length
distribution of V(D)J junctions with size peaks that are spaced by three nucleotides in
Bach2
+/+
but not Bach2
-/-
cells (Figure 5.1 B,D).
We then sequenced the IgH rearrangements (Methods Section 5.2.8) in order to
quantitatively measure and compare the percentage of non-functional rearrangements in
the bone marrow and spleen of Bach2
+/+
and Bach2
-/-
mice (Figure 5.1E, F, Figure B.1).
The results obtained were in agreement with our spectratyping results, with the knockout
bone marrow and spleen carrying a higher percentage of non-functional IgH
rearrangements as compared to the wildtype counterparts (Figure 5.1E, F, Figure B.1).
65
This experiment highlighted the requirement of Bach2 for the negative selection of non-
functional pre-B cells.
Figure 5.1. Absence of Bach2 impairs negative selection process.
A: Bone marrows were isolated from Bach2
+/+
and Bach2
-/-
mice and sorted for mature B
cells (κ
+
B220
+
). B: Spectratyping to identify the clonality status of cells sorted in part A.
C: Spleens were isolated from Bach2
+/+
and Bach2
-/-
mice and sorted for mature B cells
(κ
+
B220
+
). D: Clonality status of cells sorted in C by spectratyping. E: Percentage of B
cells with non-functional IgH rearrangements in the bone marrows of Bach2
+/+
and
Bach2
-/-
mice. F: Percentage of B cells with non-functional IgH rearrangements in the
splenic repertoire of Bach2
+/+
and Bach2
-/-
mice.

66
To test the hypothesis that Bach2 is sufficient for the clearance of pre-B cells with non-
functional IgH rearrangements, we inducibly overexpressed BACH2 in Bach2
+/+
and
Bach2
-/-
IL7-dependent pre-B cells and quantified the functional and non-functional
rearrangements in each case. BACH2 overexpression not only cleared the pool of pre-B
cells with non-functional rearrangements in Bach2
+/+
case, but also rescued completely
the negative selection in the Bach2
-/-
condition (Figure 5.2, Figure B.2).
Figure 5.2. Bach2 is both required and sufficient for negative selection of B cells
lacking functional IgH rearrangements.
IL7-dependent pre-B cells derived from bone marrows of Bach2
+/+
and Bach2
-/-
mice
were transduced with either GFP-tagged empty vector control (EV-ER
T2
) or with GFP-
tagged inducible BACH2 overexpression vector (Bach2-ER
T2
). The transduced cells
were treated with tamoxifen for 24 hours and sorted for GFP. The sequencing of V(D)J
rearrangements in the sorted cells in each case was carried out as mentioned in Section
5.2.8.

All the above experiments provided strong evidence for the absolute requirement of
Bach2 for the clearance of B cells with non-functional IgH rearrangements. In the
sections that follow, we describe the molecular players involved in BACH2-induced
negative selection at the pre-BCR checkpoint. Finally, we piece together all the
information and lay down the cell signaling pathway leading to this negative selection.

67
5.3.2. Involvement of PAX5, ARF and TP53 (p53) in BACH2-induced negative selection

The process of negative selection involves the elimination of B cells with non-functional
V(D)J rearrangements by apoptosis. As Bach2 was found to be crucial for the negative
selection process, we investigated the pathway through which this occurs. Our studies
were aimed at identifying molecules both upstream and downstream of Bach2 in this
process. To this end, we tested whether Bach2 allows clearance of non-functional pre-B
cells through the well characterized apoptosis mediators like ARF and TP53. Based on
previous reports linking PAX5 and BACH2 (Casolari et al., 2012), we hypothesized that
PAX5 may be the component upstream of Bach2 in the negative selection process.
In order to test the involvement of PAX5, ARF and TP53 in Bach2-mediated clearance of
improperly rearranged pre-B cells, we overexpressed either empty GFP vector (EV) or
Pax5-GFP vector in Bach2
+/+
and Bach2
-/-
pre-B cells, and measured the protein levels
of ARF and p53. Interestingly, empty vector-transduced Bach2
-/-
pre-B cells showed
lower basal levels of ARF and p53 as compared to Bach2
+/+
cells (Figure 5.3A).
Moreover, concordant with our hypothesis that Pax5 overexpression would fail to
upregulate ARF and p53 in the absence of Bach2, we observed that Bach2
-/-
cells do not
upregulate ARF and p53 after Pax5 overexpression, to levels comparable to the wildtype
cells (Figure 5.3A). We also carried out a growth competition assay (Methods Section
5.2.11) to compare the rate of loss of GFP
+
cells after Pax5 overexpression in Bach2
+/+

and Bach2
-/-
conditions. In contrast to the wildtype, Bach2
-/-
pre-B cells are largely
resistant to Pax5 overexpression (Figure 5.3 B-C).
The above results highlight the role of PAX5 as the key inducer of BACH2-mediated
negative selection at the pre-BCR checkpoint. They also demonstrate that BACH2
triggers apoptosis of non-functional B cells through the classical Arf-Mdm2-p53 pathway.

68
Figure 5.3. PAX5 is upstream of BACH2-induced negative selection of B cells with
non-functional IgH rearrangements.
BCR-ABL1
+
pre-B cells derived from Bach2
+/+
and Bach2
-/-
mice were transduced with
either GFP-tagged empty vector control (EV) or with GFP-tagged Pax5 overexpression
vector (Pax5-GFP) (n=3, each condition). A: Measurement of ARF and p53 levels in
each case by western blotting. B: Growth competition assay to compare growth rates of
Bach2
+/+
and Bach2
-/-
cells after Pax5 overexpression. C: Representative FACS plots
showing time-dependent change in the percentage of GFP
+
cells in each condition by
flow cytometry.

PAX5 is crucial for the V(D)J rearrangement process to form the heavy chain (μ chain) of
the pre-BCR (Rolink et al., 2000a), and represents the first checkpoint during early B cell
development. At this stage, PAX5 may transcriptionally activate Bach2 expression
(Casolari et al., 2012). We propose that, once the pre-BCR formation is complete,
BACH2 implements the next checkpoint control (i.e., negative selection) to rid the
repertoire of non-functional B cells by activating effectors like Arf and p53.

69
5.3.3. BACH2 regulates the level and activity of the RAG enzymes

Interestingly, the gene expression profiling carried out to compare Bach2
+/+
and Bach2
-/-

leukemia cells, revealed drastic downregulation of the Rag1/Rag2 recombinase
enzymes in the latter (Figure 6.14A, chapter 6). The microarray result was further
verified by qRT-PCR (Figure 5.4 A-B). In vitro differentiation of both Bach2
+/+
and Bach2
-
/-
leukemia cells using imatinib (IM) revealed drastic increase in the levels of Rag1 and
Rag2 enzymes upon treatment in the former cell type as compared to the latter (Figure
5.4 A-B). We also observed that overexpression of Bach2 in wildtye pre-B cells led to
increase in mRNA levels of Rag1 and Rag2 (Figure B.4).
Figure 5.4. BACH2 regulates level of Rag1/ Rag2 V(D)J recombinase.
A: BCR-ABL1-transformed pre-B cells derived from Bach2
+/+
and Bach2
-/-
mice were
subjected to qRT-PCR to measure mRNA levels of Rag1 before and after treatment of
imatinib (IM) (n=3, each condition). B: Rag2 mRNA levels before and after IM treatment
in Bach2
+/+
and Bach2
-/-
counterparts (n=3, each condition).

We next tested whether BACH2 also regulated the activity of the RAG enzymes. To this
end, we transduced Bach2
+/+
and Bach2
-/-
leukemia cells with the RSS-GFP reporter
construct, and carried out the assay as described in Section 5.2.9. We observed that
Bach2
-/-
cells lack the ability to rearrange a V(D)J recombination substrate (Figure 5.5 A-
B).

70
Figure 5.5. BACH2 regulates activity of RAG1/ RAG2 enzymes.
A: A diagrammatic representation of the RSS-GFP reporter construct as described in
methods section 5.2.9. B: BCR-ABL1-transformed pre-B cells derived from Bach2
+/+
and
Bach2
-/-
mice were transduced with the RSS-GFP reporter vector. The transduced cells
were selected using puromycin and subjected to an imatinib-induced in vitro
differentiation. Recombinase activity was then monitored by measuring the percentage
of GFP
+
cells in each case by flow cytometry.

Moreover, from an imatinib based in vitro differentiation assay (Section 5.2.3), we found
that Bach2
-/-
cells also fail to successfully recombine Vκ-Jκ light chain segments (Figure
5.6).
Figure 5.6. Loss of Bach2 impairs Vκ-Jκ light chain rearrangement.
Surface κ light chain staining was used to measure the proportion of cells with Vκ-Jκ
rearrangement in Bach2
+/+
and Bach2
-/-
BCR-ABL1
+
pre-B cells, before and after in vitro
differentiation using imatinib (IM).

71
We propose that BACH2 increases the chances of a cell to successfully rearrange the
heavy chain by increasing the level and activity of the RAG enzymes. In the absence of
Bach2, lower RAG activity may not only reduce the number of chances of rearrangement
but also reduce the DNA damage caused to the cell as a result of the RAG activity. Such
a reduction in the level of DNA damage may bring down the rate of negative selection
(apoptosis) of non-functional B cells. Further investigations are required to test if this
theory is indeed true.

5.3.4. BACH2 and BCL6 maintain balance between negative selection and survival of
early B cells
Having studied the function of BACH2 at the pre-BCR checkpoint, we wanted to
understand how survival of a Fraction C’ pre-B cell is triggered to allow for light chain
recombination. Previous studies from our lab have shown that BCL6 is upregulated by
the μ heavy chain (pre-BCR) (Duy et al., 2010). We have also shown that BCL6 is a
transcriptional repressor of tumor suppressors like Arf and p53, and thus promotes
survival (Duy et al., 2011). Based on these findings, we hypothesized that BACH2 (an
apoptotic inducer) and BCL6 (a pro-survival molecule) maintain the balance between
negative selection and survival respectively, at the pre-BCR checkpoint. Such a balance
is critical for a large pre-B cell to decide whether to differentiate or undergo apoptosis. It
is this balance that is upset during leukemogenesis (chapter 6).
To test our hypothesis, we analyzed how BCL6 and BACH2 are regulated at the heavy
and light chain checkpoints. As described in the previous sections 5.3.1, 5.3.2 and 5.3.3,
Bach2 is required for the process of negative selection at the heavy chain checkpoint.
From previous studies in our lab, we know that the level of BCL6 is very low prior to the
the heavy chain checkpoint (Duy et al., 2010). Therefore, BACH2 would dominate over
BCL6 at all stages prior to Fraction C’. Once Fraction C’ is reached, the cell can adopt
72
one of the two opposing pathways- apoptosis (negative selection) or survival by means
of signals emanating from the pre-BCR. During normal development, a B cell typically
checks whether recombination at IgH is successful and the PAX5-BACH2 axis would
eliminate deleterious B cells with non-functional IgH rearrangements. Pre-B cells that
cross this checkpoint successfully will upregulate BCL6 (Duy et al., 2010). We propose
that, at this stage, BCL6 would dominate over BACH2 and allow survival and further
differentiation of pre-B cells with functional IgH rearrangements. Differentiation is marked
by the rearrangement of the Vκ-Jκ segments to form the light chain (IgL). We
hypothesize that after the completion of IgL recombination, interplay between BACH2
and BCL6 would continue to influence the survival and further differentiation of Fraction
D pre-B cells. Experiments to support the above mechanism are discussed in chapter 6.
Figure 5.7. BACH2 and BCL6 maintain balance between negative selection and
survival of early B cells.
A: Illustration depicting the proposed molecular players that maintain balance between
negative selection and survival at the pre-BCR checkpoint. B: Proposed model of
competition between BACH2 and BCL6 for promoters of tumor suppressor genes like Arf
and p53 (described in detail in chapter 6).

73
5.4. Conclusions
Majority of the previous work done on Bach2 has focused on its role in mature B cells.
Lacking are the studies on Bach2’s function during the normal process of early B cell
development. As described earlier in chapter 1, it is crucial to understand the role of B
lymphoid factors like Bach2 during normal B cell development in order to understand the
mechanism of leukemogenesis. Therefore, we study Bach2 from a different dimension,
namely, its role in early B cell development.
We observe that Bach2 is essential to maintain a polyclonal B cell repertoire where
majority of the B cells that complete differentiation carry functional V(D)J
rearrangements. In the absence of Bach2, the process of negative selection is
completely abrogated, thus resulting in loss of quality control during development of B
cells.
We then dissect the mechanism of BACH2-mediated clearance of B cells with non-
functional heavy chain rearrangements. We identify PAX5, ARF and TP53 as the key
mediators of this apoptotic process. We also show that BACH2 regulates RAG1/ RAG2
recombinase activity and levels, thereby regulating the number of chances a pre-B cell
gets to carry out a successful V(D)J rearrangement. We also hypothesize that induction
of RAG1/RAG2 activity by BACH2 may impact the amount of DNA damage an early B
cell can encounter. The amount of DNA damage would then decide whether to target the
cell to survival or negative selection (apoptosis). However, further experiments need to
be carried out in this direction in order to test the credibility of this hypothesis.
Finally, we show how fine tuning of cell signaling occurs during B cell development in
order to maintain the balance between apoptosis and survival of early B cells. BACH2
and BCL6 levels are tightly regulated at every stage of B cell development depending on
whether negative selection or survival is required. The two molecules represent the yin
and yang of quality control of B cells during their early development. The study
74
described in this chapter provides us the foundation to understand the function of
BACH2 in pre-B cell leukemogenesis (ALL), which is discussed in detail in chapter 6.
5.5. Limitations and future perspectives
The above study describes the requirement of BACH2 for negative selection at the pre-
BCR checkpoint. Our study has mainly focused on the effectors of this negative
selection process downstream of BACH2. Lacking are the studies which highlight the
molecular events upstream of BACH2 which trigger this negative selection. For example,
we show that BACH2 regulates the level and activity of the RAG enzymes. It would be
particularly interesting to study whether this is a result of direct transactivation of
Rag1/Rag2 by BACH2 or an indirect effect that BACH2 has on other transcription factors
regulating RAG1 and RAG2. Hence, we are currently assessing the direct activation of
RAG1/ RAG2 promoter by BACH2 single locus ChIP in human cell lines. Experiments
also need to be carried out to check if BACH2 exposes a cell to increased DNA damage
by activating the RAG enzymes and thereby triggers DNA damage-induced cell death.
In addition to the above experiments, it would be important to dissect mechanisms other
than the Arf-Mdm2-p53 pathway in BACH2-induced negative selection. Apoptotic
processes during negative section can proceed through multiple pathways. For example,
modulation of BCL2 family members by BACH2 would shed light on whether and how
BACH2 employs mitochondrial apoptotic mechanisms during negative selection. Such a
comprehensive analysis of all the possible apoptotic pathways downstream of BACH2
was not within the scope of the current study and would be very interesting for future
investigations.

75
Chapter 6. BACH2 protects against leukemogenesis at the the pre-BCR
checkpoint
6.1. Introduction
Extensive studies have been carried out on the role of BACH2 in mature B cells and
human diffuse large B cell lymphoma (DLBCL) (discussed in Chapter 4). However, not
much is known about the role of BACH2 in pre-B ALL.
Pre-B ALL arises from deregulated cell signaling pathways in early B cell development.
We therefore, extend our understanding of the concepts learnt in chapter 5 to study the
importance of Bach2 in pre-B ALL. Using Bach2 as an example, we highlight how pre-B
ALL mirrors early B cell development in terms of the molecules and the cell signaling
pathways involved.
We and others found that BACH2 is strongly upregulated in BCR-ABL1-transformed ALL
(Ph
+
ALL) cells upon treatment with tyrosine kinase inhibitors (TKI) (Figure C.1). Recent
reports have also shown that BCR-ABL1 transcriptionally represses expression of
BACH2 by mediating PAX5, the latter being a crucial regulator of B cell differentiation
(Urbanek et al., 1994) and a frequently mutated tumor suppressor in pre-B ALL
(Mullighan et al., 2007, Medvedovic et al., 2011).
Our studies in chapter 5 show the requirement of BACH2 for clearance of non-functional
B cells. Based on this, we hypothesize that BACH2 is a guardian against pre-B
leukemogenesis. In this chapter, we dissect the tumor suppressive role of BACH2 in pre-
B ALL and its mechanism of action. We also describe in detail our studies on childhood
ALL patients which identify BACH2 as a powerful predictor of favorable clinical outcome
in children, that may be useful in future approaches for risk stratification. Our studies on
BACH2 provide a novel paradigm of leukemogenesis that arises from deregulated
signaling at the pre-BCR checkpoint.

76
6.2. Materials and Methods
6.2.1. Patient samples, human cells and cell lines
Patient samples were provided by the USC Norris Comprehensive Cancer Center in
compliance with the IRB of the University of Southern California Health Sciences
Campus. The human ALL cell lines BV173 and Tom1 were obtained from DSMZ,
Braunschweig, Germany. Human leukemia cells were maintained in Roswell Park
Memorial Institute medium (RPMI-1640, Invitrogen, Carlsbad, CA) with GlutaMAX
containing 20% fetal bovine serum, 100 IU/ml penicillin and 100 μg/ml streptomycin at
37°C in a humidified incubator with 5% CO
2
. Primary human ALL cells (patient samples)
were cultured on OP9 stroma cells.
6.2.2. Extraction of bone marrow cells from mice
Discussed in chapter 5 (Section 5.2.1).
6.2.3. In vivo model for Myc driven leukemia

IL7-dependent Bach2
+/+
and Bach2
−/−
mice were transformed with a retroviral vector
encoding Myc-IRES-GFP (Myc
GFP
). The cells were injected intrafemurally into sublethally
irradiated (250 cGy) NOD/SCID mice. Monitoring of leukemia progression in mice was
done constantly using weight loss, hunched posture and other leukemia symptoms as
indicators. After becoming terminally sick, the mice were sacrificed. Bone marrows and
spleens isolated from sacrificed mice were stained for B cell markers to confirm
leukemia as the cause of death. GFP levels were measured from the bone marrow and
spleens of the mice to account for any minimal residual disease. All mouse experiments
were subject to institutional approval by Childrens Hospital Los Angeles IACUC.
6.2.4. Bach2
+/+
, Bach2
-/-
,

Bcl6
+/+
and Bcl6
-/-
mice
Bone marrow cells from the above mentioned mice were collected and retrovirally
transformed with BCR –ABL1 in the presence of 10 ng interleukin-7 per milliliter
77
(Peprotech) in RetroNectin- (Takara) coated Petri dishes as described below. All BCR –
ABL1-transformed ALL cells derived from bone marrow of mice were maintained in
Iscove’s modified Dulbecco’s medium (IMDM, Invitrogen) with GlutaMAX containing 20%
fetal bovine serum, 100 IUml
−1
penicillin, 100 μgml
−1
streptomycin and 50 µM 2-
mercaptoethanol. BCR –ABL1-transformed ALL cells were cultured only for short periods
of time and usually not longer than for 2 months to avoid acquisition of additional genetic
lesions during long-term cell culture.
6.2.5. BCR-ABL1 Tyrosine Kinase Inhibitors (TKIs)
Imatinib was obtained from Novartis Pharmaceuticals (Basel, Switzerland) or from LC
Laboratories (Woburn, MA). TKIs were dissolved in sterile distilled water and stored
at -20ºC.
6.2.6. Western blotting
Western blotting was performed as explained in Section 3.2.5 (chapter 3). The following
antibodies were used: BACH2 (a gift from Dr. Ari Melnick’s lab, Weil Cornell Medical
College, NY), ARF (Ab80, rabbit polyclonal, Abcam) and p53 (1C12, mouse monoclonal,
Cell Signaling Technology). Antibody against β-ACTIN (C4, mouse monoclonal, Santa
Cruz) was used as a loading control.
6.2.7. Flow cytometry
Antibodies against mouse CD19 (1D3), B220 (RA3-6B2), Sca-1 (D7), CD73 (TY/23),
CD22.2 (Cy34.1), CD25 (7D4), c-kit (2B8), CD43 (S7) as well as their respective isotype
controls were purchased from BD Biosciences, San Jose, California. Antibody against
CD38 (90) was purchased from eBioscience, San Diego, California. For apoptosis
analyses, Annexin V, propidium iodide and 7-AAD (BD Biosciences) were used.
6.2.8. Colony forming assay
The methylcellulose colony forming assays were performed with 10,000 BCR-ABL1-
transformed Bach2
+/+
and Bach2
-/-
cells, or 10,000 c-myc-IRES-GFP-transformed
78
Bach2
+/+
and Bach2
-/-
IL7-dependent pre-B cells on OP9. Cells were resuspended in
murine MethoCult medium (StemCell Technologies, Vancouver, BC, Canada) and
cultured on dishes (3 cm in diameter) with an extra water supply dish to prevent
evaporation. After 7 to 14 days, colonies were counted.
6.2.9. Retroviral transduction
All retroviral transductions were carried out as described in chapter 5 (Section 5.2.6).
6.2.10. Cell-cycle analysis
For cell-cycle analysis of BCR –ABL1 ALL cells, the BrdU flow cytometry kit for cell-cycle
analysis (BD Biosciences) was used according to manufacturer’s instructions. BrdU
incorporation (APC-labelled anti-BrdU antibodies) was measured along with DNA
content (7-amino-actinomycin-D) in fixed and permeabilized cells. The analysis was
gated on viable cells that were identified based on scatter morphology.
6.2.11. Quantitative single-locus ChIP
ChIP assays were performed with modifications as described (Reynaud et al., 2008).
Briefly, 1 × 10
7
BCR-ABL1-

transformed mouse cells (Bach2
+/+
, Bach2
-/-
and Bcl6
-/-
) were
treated with or without 2 µmol l
-1
imatinib for 16 h. Then the cells were cross-linked with
1% formaldehyde. After sonication by a bioruptor (Diagenode), immunoprecipitations
were performed using 5 µg Bcl6 (C19, Santa Cruz Biotechnology) or control IgG
antibody (Santa Cruz Biotechnology). Complexes were washed with low and high salt
buffers, eluted, and reverse-crosslinked, and the DNA was precipitated.
Immunoprecipitated DNA sequences were analyzed by qRT-PCR (primer sequences
used for ChIP analyses are listed in Table C1).
6.2.12. PCR amplification and sequencing of BACH2 coding region
Total RNA from 10 primary Ph
+
ALL cases (Figure C.2, Table C2) was isolated by
RNeasy (Qiagen, Valencia, CA) purification. cDNA was generated from 5 μg of total
RNA using a poly(dT) oligonucleotide that contains a T7 RNA polymerase initiation site
79
and the SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad, CA). The entire
translated region of Bach2 was divided into 5 separate regions (A to E) for amplification
by PCR and sequencing. PCR was performed with Phusion polymerase (New England
BioLabs Inc.) for 35 cycles using the template cDNA derived from the 10 primary cases.
Sequences of primers used for PCR and sequencing are provided in Table C1.
6.2.13. Sequence alignment to identify mutations
Sequencing was carried out by USC Norris Comprehensive Cancer Center DNA core
facility. All sequences were aligned to the wildtype BACH2 cDNA sequence (obtained
from Ensemble Genome Browser) in Bioedit, and were compared for mutations. In
addition, mutations were identified by carefully reading through the sequencing
chromatograms to identify double peaks at a particular position in the trace. Sequence
data are available from EMBL/GenBank under accession numbers HE578168,
HE578164, HE578169, HE578170, HE578165, HE578163, HE578173, HE578174,
HE578166, HE578176.
6.2.14. ROS staining
For evaluation of intracellular ROS levels, Bach2
+/+
and Bach2
-/-
ALL cells were
incubated for 7 min with 1 µM 5-(and 6-)chloromethyl-2′,7′-dichlorodihydrofluorescein
diacetate (CM-H
2
DCFDA, Invitrogen, Carlsbad, CA) at 37°C for oxidation of the dye by
ROS. After washing with PBS, the cells were incubated an additional 15 minutes at 37°C
in PBS to allow complete deacetylation of the oxidized form of CM-H
2
DCFDA by
intracellular esterases. The levels of fluorescence were then directly analyzed by flow
cytometry, gated on viable, PI
-
cells.
6.2.15. Affymetrix GeneChip analysis
Total RNA from cells used for microarray or RT-PCR analysis was isolated by RNeasy
(Qiagen, Valencia, CA) purification. RNA quality was first checked by using an Agilent
Bioanalyzer (Agilent Technologies, Santa Clara, CA). cDNA was generated from 5 μg of
80
total RNA using a poly(dT) oligonucleotide that contains a T7 RNA polymerase initiation
site and the SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad, CA).
Biotinylated cRNA was generated and fragmented according to the Affymetrix protocol
and hybridized to U133A 2.0 human or 430 mouse microarrays (Affymetrix, High
Wycombe, UK). After scanning (scanner from Affymetrix) of the GeneChip arrays, the
generated CEL files were imported to BRB Array Tool (http://linus.nci.nih.gov/BRB-
ArrayTools.html) and processed using the RMA algorithm (Robust Multi-array Average)
for normalization and summarization. Relative signal intensities of probe sets were
determined by comparing the intensity of treated or untreated cells to the average value
of the cell line or a group of cells. Ratios were exported in Gene Cluster and visualized
as a heatmap with Java Treeview. Microarray data are available from GEO under
accession numbers GSE34861, GSE34937, GSE30883, GSE24814, GSE30928,
GSE21664, GSE20987.
6.2.16. BACH2 gene expression data and clinical outcome
RMA data for the 207 P9906 patients was obtained from the TARGET web site
(target.cancer.gov/dataportal/data/; ALL RMA Probeset-Level Expression). Three probe
sets were available for BACH2 (221234_s_at, 227173_s_at, 236796_at). For various
analyses, each probe set was assessed separately, or alternatively, since the
expression patterns for the multiple probe sets were so similar, the mean value of the
intensities were calculated for each gene and used for subsequent analysis.
None of the three probe sets were identified as “outliers” in our unsupervised analysis.
However, BACH2 was among the top rank order probe sets associated with cluster R3
and was also listed among the top down-regulated (very low levels of expression)
signature genes in BCR-ABL1 patients. We also evaluated the expression of BACH2 as
they related to the other known clinical variables (age, sex, translocations, WBC, MRD,
ethnicity, +4/+10 and Down Syndrome; DS). There were no significant correlations found
81
with age, sex, WBC, DS or ethnicity. The E2A-PBX1 translocations (cluster R2) and a
related cluster (R3) were highly correlated with the expression of BACH2.

6.3. Results
6.3.1. Transcriptional inactivation of BACH2 in ALL
Pre-B ALL patients were categorized into sub-groups based on the chromosomal
translocations they carried, and their gene expression profiles were compared to pre-B
cells from normal human counterparts. On doing so, it was observed that BACH2 mRNA
levels were significantly lower in high risk ALL groups, namely the BCR-ABL1
(p=0.00203) and MLL-AF4 (p=0.01106) sub-groups (Figure 6.1A).
On comparing the methylation status of BACH2 promoter in BCR-ABL1 patients (n=84)
with normal human counterparts (n=12), we found that BACH2 promoter was
significantly hypermethylated (p= 0.0026) in the pre-B cells of Ph
+
ALL patients as
compared to normal pre-B cells (Figure 6.1B).
A detailed sequence analysis of the BACH2 coding region in 10 primary cases of
Ph
+
ALL revealed 7 unique point mutations including 5 amino acid changes in the
BACH2 BTB domain (Figure C.2, Table C2). These findings suggest that BACH2 is
affected by somatic mutations in a fraction of Ph
+
ALL cases.

82
Figure 6.1. Human Ph
+
ALLs are characterized by low BACH2 expression levels.
A: Gene expression values of BACH2 in different subgroups of pre-B ALL patients (van
Zelm et al., 2005; Ross et al., 2003) B: HELP assay carried out to measure genome-
wide methylation status in 12 normal and 84 Ph
+
ALL patients showed increased
methylation at the BACH2 locus in ALL patients in comparison to normal counterparts.
(HELP assay performed in Melnick Lab).

An independent study compared the gene expression profiles at diagnosis and relapse
in 49 pairs of childhood ALL patients. In 44 of these pairs, the relapse samples showed
drastically reduced mRNA levels of BACH2 (p=0.019), suggesting that loss of BACH2
expression is associated with relapse of childhood ALL (Figure 6.2). Upon further
classification of the relapse patients into early and late categories, an identical pattern to
the one described above was observed (Figure 6.2).

83
Figure 6.2. B-ALL patients show decreased BACH2 expression levels upon
relapse.
Gene expression signature of 49 childhood ALL patients was determined at the time of
diagnosis and again at relapse using microarrays (Hogan et al, 2011). Patients were
classified into two groups (early and late) depending on how soon they relapsed. A
meta-analysis was carried out to compare and contrast BACH2 gene expression levels
at the time of diagnosis and relapse in each patient.

The findings in Figure 6.2 are consistent with copy-number variation (CNV) analyses that
identified small deletions at 6q15 in 4 out of 11 cases of relapsed childhood ALL. All
these deletions at 6q15 encompassed the BACH2 locus (Figure 6.3 A-E).
Figure 6.3. Deletions on chromosome 6 involving BACH2 locus in relapse ALL
patients.
A: Scatter plots of log
2
ratio value for each probe mapping to chromosome 6 from the
Affymetrix 500K Array Set for the matched diagnostic and relapse samples from four
patients (NCL45, NCL405, NCL578, NCL625). Deletions encompassing BACH2 are
indicated by grey bars. B: Scatter plot of the focal deletion on 6q15 in the relapse (R) as
compared to the diagnostic (D) sample of patient NCL625 shows that the deletion
encompasses BACH2. C: Copy number real-time PCR of BACH2 confirms the
microarray data indicating that both alleles are deleted in the relapse samples of
NCL405 and NCL625. The bar chart shows copy number for the diagnostic (dark grey),
remission (dark green) and relapse (dark red) for three patients as compared to the
normal counterpart. This analysis also confirms that there is deletion at both diagnosis
and relapse in patient NCL578 as detected by microarray analysis. D: Bar chart of the
TaqMan real-time PCR analysis of BACH2 (Hs00222364_m1) shows a significant
reduction in the gene expression levels of BACH2 at relapse (dark red) compared with
diagnosis (dark grey) (P-value = 0.0027). No significant difference in expression was
observed for patient NCL578. E: Western blot analysis to determine levels of protein
expression in patient NCL405 (relapse sample only) and the diagnostic (D) and relapse
84
(R) samples from patients NCL625 and NCL578. No expression of BACH2 in the relapse
sample of patient NCL405 was seen. Reduced expression was observed at relapse as
compared with diagnosis in patients NCL625 and NCL578.

85
Figure 6.3 continued.

Numerous studies carried out in collaboration with Children’s Oncology Group
(NCT00005603 COG P9906), revealed the importance of BACH2 as a powerful
predictor of favorable clinical outcome in children. Patients who expressed higher levels
of BACH2 had a longer relapse free survival rate than patients who expressed lower
levels of BACH2 (n=207; p<0.03) (Figure 6.4A). Multivariate analysis of BACH2 levels
versus MRD (Minimal Residual Disease) positivity showed that patients with low BACH2
levels and positive MRD have the poorest clinical outcome (Figure 6.4B).
86
Figure 6.4. High BACH2 expression levels are an indicator of favorable outcome in
B-ALL patients.
A: 207 childhood ALL patients were divided into 2 categories (high and low) based on
their BACH2 expression levels. Relapse free survival (RFS) was compared between the
two groups by Kaplan Meier Analysis. B: Multivariate analysis of BACH2 mRNA levels
versus minimal residual disease (MRD) on day 29 to compare RFS in the pediatric ALL
samples. (COG, P9906).

BACH2 mRNA levels in childhood ALL samples at diagnosis negatively correlated with
early minimal residual disease (MRD) findings on day 29 (p<0.0001) (Figure 6.5A). A
stratification of these patients into low, intermediate and high risk categories showed that
high expression levels of BACH2 were associated with the low risk ALL group (Figure
6.5B).
Other studies carried out as a part of the same clinical trial revealed that most kinase
driven (Ph-like) leukemia had lower BACH2 levels (Figure 6.5C), corroborating our
previous findings in Figure 6.1A-B. In addition, patients with low BACH2 levels fell into
the worst outcome R8 (Figure 6.5D) group of the ROSE cluster (Harvey et al., 2010).
All the findings in childhood ALL patients point to BACH2 as a putative tumor-suppressor
in pre-B ALL. Therefore, the following experiments in this chapter have been aimed at
addressing the mechanisms of BACH2-induced leukemia suppression.

87
Figure 6.5. Pediatric ALL patients with low BACH2 expression levels have poor
overall prognosis.
A: 207 ALL patients were classified into two categories based on their MRD status on
day 29 and their patterns of BACH2 expression were mapped and compared. B: BACH2
expression levels could be successfully used to segregate patients in the low risk
category from intermediate and high risk groups. C: Patients were stratified based on
whether their leukemia possessed a kinase signature (Ph
+
) or not (Ph
-
) and the patterns
of BACH2 expression were compared between the two groups. D: ROSE clustering was
used to divide patients into 8 groups (R1-R8), R8 being the group with the worst
prognosis. Patients in the R8 group (R8
+
) possessed lower BACH2 expression levels in
comparison to groups R1-R7 (R8
-
). For all Figures 6.5 A-D, BACH2 expression levels
are displayed along the x-axis. (COG, P9906).

88
6.3.2. BACH2 inhibits oncogenic transformation of pre-B cells

In this section, we study the role of BACH2 in inhibiting the transformation of a pre-B cell
by oncogenes like c-Myc, BCR-ABL1 and constitutively active Stat5 (STAT5
CA
). Our
studies have been carried out using pre-B cells isolated from Bach2
-/-
mice and their
wildtype counterparts.
IL7-dependent pre-B cells from Bach2
+/+
and Bach2
-/-
mice were transduced with BCR-
ABL1 and proliferation, self renewal and the percentage of apoptosis were compared.
Upon transformation with BCR-ABL1, Bach2
-/-
cells showed higher proliferation (Figure
6.6A), self-renewal (Figure 6.6B) and lower spontaneous apoptosis (Figure 6.6C) as
compared to wildtype counterparts.
Ph
+
leukemia patients are treated with the BCR-ABL1 kinase inhibitor imatinib
(commonly known as gleevec). Hence, we wanted to test if loss of Bach2 hindered the
response of BCR-ABL1-transformed mouse pre-B cells to imatinib. In Bach2
+/+
leukemia
cells, BCR-ABL1 kinase inhibition with Imatinib results in rapid induction of cell cycle
arrest (Figure 6.6A) and apoptosis (Figure 6.6C), as expected. By contrast Bach2
-/-
cells
are largely resistant to Imatinib (Figure 6.6A,C), suggesting that loss of BACH2 function
in Ph
+
ALL may contribute to Imatinib-resistance in Ph
+
ALL patients.

89

Figure 6.6. Bach2 slows down BCR-ABL1-mediated transformation of pre-B cells.
A: BrdU assay was used to measure the differential proliferation rates of Bach2
+/+
and
Bach2
-/-
cells after transformation with BCR-ABL1. B: Colony forming abilities of the
Bach2
+/+
and Bach2
-/-
BCR-ABL1- transformed cells were compared using the standard
methylcellulose colony formation assay. C: AnnexinV/7AAD double staining was
employed to compare the percentages of BCR-ABL1
+
leukemia cells undergoing both
spontaneous and Imatinib-induced apoptosis in Bach2
+/+
and Bach2
-/-
counterparts.

We next studied the effect of Myc overexpression on Bach2
+/+
and Bach2
-/-
BCR-ABL1-
transformed cells. Retroviral overexpression of Myc induced apoptosis in Bach2
+/+

leukemia cells. However, overexpression of Myc had no significant effect in Bach2
-/-

leukemia cells (Figure 6.7A). Cell cycle analysis revealed that overexpression of Myc in
Bach2
+/+
ALL cells increased the fraction of cells in S-phase to a maximum threshold of
about 50%. By contrast, Myc overexpression in Bach2
-/-
ALL cells caused a drastic
increase of proliferation with about 70% of pre-B cells in S-phase (Figure 6.7B).

90
Figure 6.7. Bach2 is required for MYC-induced apoptosis.
A: BCR-ABL1 transformed Bach2
+/+
and Bach2
-/-
pre-B cells were transduced with either
control vector (EV) or Myc overexpression vector following which percentage of
apoptosis (AnnexinV
+
/7AAD
+
) was measured in each case. B: Myc- transduced Bach2
+/+

and

Bach2
-/-
leukemia cells were subjected to BrdU staining to compare the percentages
of cells in S phase in each condition.

MYC-driven proliferation of pre-B cells cannot be increased beyond a certain limit which
is defined by failsafe mechanisms. Aberrant proliferation beyond these limits will activate
apoptosis and terminate pre-malignant clones before they can acquire additional genetic
lesions (Eischen et al., 1999; Post et al., 2010). Here we show that forced expression of
Myc leads to apoptotic cell death in Bach2
+/+
IL7- dependent pre-B cells (Figure 6.8 A-D).
In Bach2
-/-
pre-B cells, however, forced expression of Myc leads to leukemic
transformation (Figure 6.8 A-D). In the absence of Bach2, failsafe mechanisms and
apoptosis are not activated, obviating the requirement of additional genetic lesions for
Myc-driven leukemic transformation.
Figure 6.8. Bach2 prevents leukemic transformation by Myc in vitro.
A: Bach2
+/+
and Bach2
-/-
IL7-dependent pre-B cells were transduced with Myc-GFP and
their rates of transformation were compared by measuring the change in percentage of
GFP positive cells with time. B: Colony formation assay to compare self-renewal
potentials of Myc- transduced Bach2
+/+
and Bach2
-/-
IL-7 dependent pre-B cells. C:
91
Comparison of the size of early apoptotic population (AnnexinV
+
/ 7AAD
-
) in each case
after Myc transduction. D: AnnexinV/7AAD staining comparing the percentage of late
apoptotic and dead cells (AnnexinV
+
/ 7AAD
+
) in the above mentioned conditions.

To confirm the critical role of Bach2 in Myc-induced failsafe control in an in vivo setting,
IL7-dependent Bach2
+/+
and Bach2
-/-
pre-B cells were transduced with a retroviral Myc
GFP vector. Following this, Myc
GFP+
pre-B cells were sorted and injected into sublethally
irradiated NOD/SCID recipient mice. While Bach2
-/-
Myc
GFP
pre-B cells readily initiated
fatal leukemia in NOD/SCID recipient mice (all mice in this group died within 25 days), all
NOD/SCID mice receiving Bach2
+/+
Myc
GFP
pre-B cells were still healthy at 70 days after
injection (p<0.0001; Figure 6.9A-C). After 70 days, the mice in the surviving Bach2
+/+

Myc
GFP
cohort were sacrificed and analyzed together with mice which had received
Bach2
-/-
Myc
GFP
pre-B cells. In contrast to mice injected with Bach2
-/-
Myc
GFP
pre-B cells,
92
no CD19
+
GFP
+
cells (i.e. leukemia) were detectable by flow cytometry in the bone
marrow and spleen of the group which received the Bach2
+/+
Myc
GFP
pre-B cells. MRD
PCR for leukemic cells using GFP-specific primers revealed small numbers of persisting
leukemia cells in mice which received Bach2
+/+
Myc
GFP
pre-B cells. However, MRD levels
were 4-5 log
10
orders lower than mice injected with Bach2
-/-
Myc
GFP
pre-B cells (Figure
6.9C). The above results demonstrate that BACH2 is a failsafe barrier that prevents
Myc-induced leukemic transformation of a pre-B cell.
Figure 6.9. Bach2 prevents leukemic transformation by Myc in vivo.
A: Bach2
+/+
and Bach2
-/-
IL7-dependent pre-B cells were transduced with Myc GFP and
100,000 cells were injected into 7 immune-deficient NOD-SCID mice per group.
Experiment was carried out as discussed in Methods Section 6.2.3. B: Percentages of
CD19
+
/GFP
+
cells in the bone marrow and spleen of mice from both groups were
measured by flow cytometry to ensure that B cell leukemia was the cause of death. C:
GFP levels were measured by qRT-PCR using bone marrow and spleen isolated from
each of the 7 mice in both groups to determine if there was any hidden MRD in the
Bach2
+/+
group.

93
Like MYC, hyperactive STAT5 (Stat5
CA
) can also induce apoptosis (Nosaka et al., 1999).
Here we show that Bach2
-/-
but not Bach2
+/+
pre-B cells are permissive to the
overexpression of constitutively active Stat5 (Stat5
CA
; Figure 6.10A-B). Together, the
findings in Section 6.3.2 establish that BACH2 inhibits oncogenic transformation of pre-B
cells by inducing apoptosis.
Figure 6.10. Bach2 prevents leukemic transformation by Stat5
CA
.
Bach2
+/+
and Bach2
-/-
IL7-dependent pre-B cells were transduced with empty GFP vector
(EV) or with Stat5
CA
-GFP (Stat5
CA
) vector. Growth competition assay was carried out to
measure the change in percentage of GFP
+
cells with time. A: Representative flow
cytometry of days 0 and 4 after transduction highlighting the percentage of GFP
+
cells in
each case. B: Growth competition assay comparing the change in percentage of GFP
+

cells with time across the four conditions.

BACH2 is critical in restraining Myc expression levels and limits accumulation of BCR-
ABL1-induced ROS (reactive oxygen species) production. Bach2
-/-
ALL cells were
permissive to significantly higher levels of Myc overexpression (Figure 6.11A) and were
able to survive much higher levels of ROS (Figure 6.11B). Previous work demonstrated
94
that Myc-induced ROS accumulation results in DNA damage and genetic instability
(Vafa et al., 2002). Hence, early loss of Bach2 function may play an important role in
accelerating the evolution of Myc-driven pre-leukemic pre-B cells.
Figure 6.11. BACH2 limits permissiveness of pre-B cells to Myc expression levels
and to ROS accumulation.
Bach2
+/+
and Bach2
-/-
pre-B cells were challenged with overexpression of Myc. A:
Histogram highlighting different fluorescence intensities of GFP following Myc-GFP
transduction of Bach2
+/+
and Bach2
-/-
cells. B: Consistent with the finding that Bach2
-/-

cells are permissive to higher levels of Myc expression, these cells can also accumulate
higher levels of ROS downstream of BCR-ABL1 signaling. ROS levels were visualized
by staining with the DCF substrate and measured by flow cytometry.

Based on the above findings, we hypothesized that loss of Bach2 expression may occur
during progressive leukemic transformation. This possibility was directly tested in pre-B
cells from BCR-ABL1-transgenic mice (Heisterkamp et al., 1990) and wildtype controls.
Pre-B cells were isolated from the bone marrow of BCR-ABL1-transgenic mice at
various ages and compared to wildtype controls, and mice with leukemia after treatment
with the tyrosine kinase inhibitor Nilotinib

(Trageser et al., 2009). Around 90 days of age,
BCR-ABL1-transgenic mice develop overt leukemia. These mice are phenotypically
normal before the onset of leukemic transformation. We observed that Bach2 expression
levels progressively decrease during the leukemic transformation of a pre-B cell. At the
95
same time, expression levels of Myc progressively increase together with Arf/p53
mediators of oncogenic stress (Figure 6.12).
Figure 6.12. Loss of Bach2 during progressive leukemic transformation.
Pre-B cells in BCR-ABL1-transgenic mice are phenotypically indistinguishable until
approximately 30 days after birth and subsequently undergo progressive transformation
by BCR-ABL1. In this analysis, normalized gene expression values for Bach2, Bcl6,
Myc, p53 and Arf (Cdkn2a) are plotted against the age of mice [days], reflecting the
stage of the transformation process. Wildtype mice were set as “0 days’, BCR-ABL1-
transgenic mice that recovered from leukemia after treatment with Nilotinib, were set at
“150 days”. (Trageser et al., 2009).

6.3.3. Mechanisms of BACH2- induced tumor suppression
In previous sections, we describe studies that point to Bach2 as failsafe barrier against
pre-B leukemogenesis. In the current section, we dissect the mechanisms by which
Bach2 mediates this effect. Our findings from chapter 5 (Section 5.3.4) implied that
relative levels of BACH2 and BCL6 would decide between apoptosis or leukemic
transformation of a pre-B cell in response to oncogenic stimuli. In the following section,
we provide experimental evidence to back our hypothesis.
The first and foremost goal was to test if BACH2 induced apoptosis through the classical
Arf-p53-Mdm2 tumor suppressor pathway. To this end, we compared the protein levels
96
of ARF and TP53 in Bach2
+/+
and Bach2
-/-
BCR-ABL1-transformed cells alongside with
Bcl6
+/+
and Bcl6
-/-
BCR-ABL1 cells. Protein levels of both ARF and TP53 were lower in
Bach2
-/-
cells as compared to their wildtype counterparts. As expected, this pattern was
exactly opposite to that observed in Bcl6
+/+
and Bcl6
-/-
BCR-ABL1
+
cells (Figure 6.13A).
When Bach2
+/+
and Bach2
-/-
BCR-ABL1
+
cells were transduced with Myc, ARF and TP53
were upregulated to a much higher extent in the Bach2
+/+
cells in comparison to the
knockout counterparts (Figure 6.13B). This corroborated our findings in Figure 6.7A-B
where Bach2
-/-
cells showed reduced apoptosis and increased proliferation after Myc
overexpression in comparison to Bach2
+/+
wildtype counterparts.
Figure 6.13. Bach2 inhibits leukemic transformation by activating the classical Arf-
Mdm2-p53 tumor suppressor pathway.
A: BCR-ABL1 transformed Bach2
+/+
and Bach2
-/-
pre-B cells and Bcl6
+/+
and Bcl6
-/-
pre-B
cells were compared side by side for expression levels of ARF and p53 proteins by
western blotting (n=3, each case). B: Bach2
+/+
and Bach2
-/-
BCR-ABL1 leukemia cells
were transduced with either empty vector (EV) or Myc overexpression vector, following
which protein levels of ARF and p53 were measured by western blotting (n=3, each
condition).

The results described above supported our notion that BACH2 inhibited oncogene-
induced leukemic transformation by activating classical tumor suppressors like ARF and
p53.
A complete gene expression profiling of the BCR-ABL1- transformed Bach2
+/+
and
Bach2
-/-
cells was carried out to compare and characterize the two cell types (Figure
6.14 A-B). The profile showed downregulation of a number of tumor suppressor genes
97
like Arf (Cdkn2a) and Btg2 (Figure 6.14A), upon loss of Bach2 expression. These
studies are in agreement with our earlier findings in Figure 6.13.
Figure 6.14. Phenotype of Bach2-deficient acute lymphoblastic leukemia cells.
A: Gene expression profiling of BCR-ABL1 transformed Bach2
+/+
and Bach2
-/-
pre-B cells
using microarrays (n=3). B: Verification of the phenotype of Bach2
+/+
and Bach2
-/-

leukemia cells by flow cytometry.

A systematic analysis revealed that 134 of the 541 and 565 BCL6- and BACH2-target
genes are shared, respectively (Figure 6.15B). For the majority of shared BCL6/BACH2-
target genes, BCL6 and BACH2 affect transcriptional activity in the same direction. For a
small group of tumor-suppressor genes which include Cdkn2a (Arf) and Btg2, however,
BCL6 and BACH2 have opposite effects (Figure 6.15A). Since BCL6 functions as a
98
strong transcriptional repressor of these loci in normal pre-B cells

(Duy et al., 2010) and
Ph
+
ALL (Duy et al., 2011), we hypothesized that BACH2 interferes with BCL6-mediated
transcriptional repression of Arf/p53.
Figure 6.15. BACH2 reverses BCL6-dependent gene expression changes.
A: Gene expression profile of BCR-ABL1 transformed Bach2
+/+
and Bach2
-/-
pre-B cells
as compared to Bcl6
+/+
and Bcl6
-/-
leukemia cells (only genes which demonstrate an anti-
correlation relationship between the two signatures are shown). B: Venn diagram
highlighting genes which are differentially regulated by both BACH2 and BCL6.

Binding of both BCL6 and BACH2 to CDKN2A and TP53 promoters affects gene
expression, although in opposite ways: Protein levels of ARF and p53 are significantly
reduced in the absence of BACH2 but strongly increased in the absence of BCL6
(Figure 6.13A). Likewise, mRNA levels of the p53-dependent tumor suppressor Btg2
(Rouault et al., 1996) were reduced by >20-fold in Bach2
-/-
leukemia cells (Figure C.3). It
was plausible that the contrast in outcome observed upon binding of BACH2 and BCL6
to the promoters of the same tumor suppressor genes was purely coincidental. However,
99
two facts argued against this - 1) BACH2 and BCL6 are both BTB domain-containing
proteins which could interact and thereby decide the fate of transcription at identical
target genes and 2) Relative levels of BACH2 and BCL6 are tightly regulated at every
stage of B-cell development to maintain balance between negative selection and survival
(chapter 5). Hence, it was logical to hypothesize that BACH2 and BCL6 shared an
antagonistic relationship in leukemogenesis, identical to the one seen in early B cell
development (Figure 6.16). Shown below is the proposed model for the antagonistic
relationship between BACH2 and BCL6.
Figure 6.16. Bach2 and BCL6 maintain balance between oncogene-induced
apoptosis and leukemic transformation of early B cells.
A: Diagram illustrating the key molecular players that decide between apoptosis and
leukemogenesis at the pre-BCR checkpoint. B: Proposed model of competition between
BACH2 and BCL6 for promoters of tumor suppressor genes like Arf and p53.

Of note, ChIP-seq analysis in a human B-lymphoma line revealed that BCL6 and BACH2
share binding sites on the promoters of CDKN2A (Arf), TP53 and number of other tumor
suppressor genes (Figure 6.17). This result prompted us to carry on subsequent studies
which shed light on the antagonistic relationship between the two proteins in leukemia.
100
Figure 6.17. BACH2 and BCL6 share binding sites on promoters of well-known
tumor suppressor genes.
ChIP sequencing analysis of human lymphoma cells reveals that BACH2 and BCL6
have common binding sites. Shown in this figure are some of the common tumor
suppressor genes which are targets of both BACH2 and BCL6. As depicted in the ChIP
sequencing tracks, the binding peaks of BACH2 and BCL6 overlap for most of these
genes, indicating a possible competition between the two proteins for binding to these
sites. (Experiment performed by Chuanxin Huang, Melnick Lab).

To directly test the hypothesis that BACH2 negatively regulates the ability of BCL6 to
repress CDKN2A (Arf) and TP53 (p53), we performed BCL6-ChIP experiments with
101
Bcl6
-/-
(negative control), Bach2
+/+
and Bach2
-/-
cells (Figure 6.18A). Increased binding of
BCL6 to Arf and p53 promoters was observed in the absence of Bach2 (Figure 6.18A).
We confirmed that this was not a result of increased Bcl6 expression in Bach2
-/-
ALL
cells (Figure 6.18B). These findings indicate that relative levels of BCL6 and BACH2
decide whether repression or transcriptional activation would occur at CDKN2A (Arf) and
TP53 (p53) loci.
Figure 6.18. BACH2 inhibits recruitment of BCL6 to CDKN2A (Arf) and TP53 (p53)
promoters.
A: BCL6-ChIP was performed on BCR-ABL1 transformed Bach2
+/+
and Bach2
-/-
cells to
check if Bach2 status was a determinant of the amount of BCL6 recruited to Arf and p53
promoters. B: qRT-PCR to confirm that increased binding of BCL6 to Arf and p53
promoters seen in Bach2
-/-
cells is not a result of increased BCL6 levels in the Bach2
-/-

cells (n=3, each condition).

102
An independent experiment which compared gene expression profiles of Bcl6
+/+
and
Bcl6
-/-
leukemia cells after inducible over-expression of BACH2, revealed increased
upregulation of Arf, p53 and Btg2 only in Bcl6
-/-
cells but not in their wildtype
counterparts (Figure 6.19). These results further strengthen the idea that BCL6 and
BACH2 inhibit each other’s binding to promoters of common target genes like Arf and
p53.
Figure 6.19. BCL6 reverses BACH2-mediated transcriptional activation of Arf/p53.
BACH2-induced activation of tumor suppressor genes like Arf, p53 and Btg2 is
dependent on BCL6 genotype. A: Gene expression profiling carried out after inducible
BACH2 overexpression in Bcl6
+/+
and Bcl6
-/-
cells. B: qRT-PCR for Arf mRNA to verify
the microarray result in part A. C: qRT-PCR for p53 mRNA using samples generated in
A.

103
The previously described results provide compelling evidence for the balance between
BACH2 and BCL6 as a key factor regulating the decision of a leukemic cell to proliferate
or undergo apoptosis. Such a balance is similar to the one seen between negative
selection and survival at the pre-B cell receptor checkpoint during early development
(chapter 5). Concordant with these observations, we observed that patients who had
high BACH2 and low BCL6 had the highest relapse free survival probability and patients
with low BACH2 and high BCL6 had the lowest relapse free survival probability (Figure
6.20).
Figure 6.20. Relapse free survival probability of pediatric ALL patients is
dependent on the relative levels of BACH2 and BCL6.
Multivariate analysis of BACH2 versus BCL6 mRNA levels to compare RFS in the
pediatric ALL samples. (COG, P9906)

6.3.4. BACH2-induced apoptosis is dependent on p53 status
Experiments carried out in previous sections highlight the involvement of the Arf-Mdm2-
p53 pathway in BACH2-mediated tumor suppression. However, whether Bach2
mediates this tumor suppression through both ARF and p53 or solely through one of
these molecules remained to be tested. To this end, Arf
+/+
p53
+/+
, Arf
-/-
and p53
-/-
BCR-
ABL1- transformed leukemia cells were transduced with either empty GFP vector control
104
(EV) or with a vector over-expressing Bach2-GFP. Change in percentage of GFP
+
cells
with time was measured using a growth competition assay (described in Materials and
Methods, Chapter 5). Bach2 overexpression led to rapid elimination of GFP
+
cells in the
Arf
+/+
p53
+/+
and Arf
-/-
conditions. However, the p53
-/-
BCR-ABL1 cells were largely
resistant to Bach2 over-expression (Figure 6.21A-B). We thus infer that p53 is an
absolute requirement for BACH2-induced apoptosis.
Figure 6.21. Bach2-induced tumor suppression is p53-dependent.
A: Representative flow cytometry plots comparing the change in percentage of GFP
+

cells after overexpression of Bach2/GFP or empty vector (EV) in BCR-ABL1 transformed
Arf
+/+
p53
+/+
, Arf
-/-
and p53
-/-
pre-B cells. B: Growth competition assay depicting the
increased resistance of only the p53
-/-
BCR-ABL1 transformed cells to Bach2
overexpression in comparison to their p53
+/+
wildtype counterparts.

105
In order to test the relevance of Arf/p53 status on Bach2-induced tumor suppression in a
human leukemia setting, we extended our study to primary patient derived Ph
+
ALL
samples. First, we carried out a western blot to obtain the expression profile of ARF and
TP53 in a panel of 10 Ph
+
ALL cases (Figure 6.22A). All our cases showed TP53
expression, though some lacked ARF expression. We selected three out of these ten
cases and carried out a growth competition assay after expression of empty GFP vector
(EV) or Bach2-GFP overexpression vector. In all three cases, Bach2 overexpression led
to rapid loss of GFP
+
cells with time (Figure 6.22B-C), confirming our previous
experimental result which highlighted the importance of p53 in BACH2-induced tumor
suppression (Figure 6.21). Based on this, one would predict that human leukemia which
lack p53 expression would be more resistant to chemotherapy agents which rely on
BACH2 for inducing cell death.
Figure 6.22. Overexpression of Bach2 in primary Ph
+
ALL cells induces drastic cell
death by a p53-dependent mechanism.
A: Expression status of p53 and ARF proteins in 10 Ph
+
ALL cases. Three Ph
+
cases
(LAX2, TXL3 and SFO2) were selected from A and the effect of Bach2 overexpression in
them was studied. B: Representative FACS analysis of days 1 and 18 after transduction,
depicting the percentage of GFP
+
cells in each case. C: Growth competition curve
showing reduction in percentage of GFP
+
cells in all three primary Ph
+
ALL cases upon
overexpression of Bach2-GFP.

106
Figure 6.22. continued.

107
6.3.5. BACH2 as an effector in PAX5-mediated tumor suppression in ALL

Recent reports have shown that BCR-ABL1 transcriptionally suppresses BACH2 by
mediating PAX5 (Casolari et al., 2012). Interestingly, our analysis of leukemia carrying
dominant negative PAX5, namely, TEL-PAX5 and PAX5-C20ORF112 (Kawamata et al.,
2012) showed a drastic downregulation of BACH2 expression in comparison to leukemia
samples with wildtype PAX5 (Figure 6.23B-C). Moreover, these dominant negative
versions of PAX5 failed to activate transcription of BACH2 in a luciferase reporter assay
to the same extent as wildtype PAX5 (Figure 6.23A). The above results reveal the
importance of BACH2 as one of the downstream effectors of PAX5-induced checkpoint
control in Ph
+
ALL.
Figure 6.23. Induction of BACH2 expression is compromised in ALLs containing
PAX5 fusions.
A: Luciferase reporter assay to measure extent of activation of BACH2 promoter after
transduction of pre-B cells with empty, PAX5 overexpression or PAX5-ETV6 vectors. B:
Downregulation of BACH2 expression by both dominant negative versions of PAX5 as
measured by microarrays. C: Verification of microarray result shown in B by qRT-PCR.
(Experiment performed by Koeffler Lab, UCLA).

108
6.4. Conclusions
BACH2 is a B-lymphoid specific transcription factor of the bZip family containing a BTB
domain (Oyake et al., 1996), which has been characterized as a tumor suppressor in
diffuse large B cell lymphoma (Sakane-Ishikawa et al., 2005). In the previous chapter,
we identified BACH2 as an absolute requirement for negative selection of non-functional
B cells at the pre-B cell receptor checkpoint. From the results described in chapter 5, it
logically followed that disruption of Bach2 would eliminate one of the key guardians of
checkpoint signaling during B cell development, and as a consequence, increase the
propensity of a pre-B cell to undergo leukemogenesis. Therefore, in this chapter we
focus on the role of BACH2 in pre-B ALL.
We begin our study with an in-depth analysis of BACH2 status in human ALL patients,
enrolled in different studies. We identify that high BACH2 levels are associated with
favorable outcome, negative MRD and lower risk in these patients. We also find that
BACH2 levels are reduced when patients develop relapsed ALL and that this could
result from the acquisition of large chromosomal deletions involving the BACH2 locus.
The clinical data obtained from all these studies were in perfect agreement with one
another and pointed to BACH2 as a putative tumor suppressor in pre-B ALL.
Next, using a knockout mouse model of Bach2, we verify the tumor suppressor role of
Bach2. In addition, we identify that Bach2 acts as a barrier against pre-B cell
leukemogenesis induced by oncogenes like BCR-ABL1, cMYC and STAT5. This is
particularly interesting in the case of cMYC because all reports thus far highlight the
resistance of a pre-B cell to Myc-induced transformation (Eischen et al., 1999; Post et al.,
2010). However, in the absence of Bach2, Myc-induced transformation of IL7-dependent
pre-B cells not only occurs faster but results in generation of full-blown leukemia when
the cells are injected into immune-deficient mice. All these results highlight the
importance of Bach2 as a guardian against pre-B cell leukemogenesis.
109
Following this, we dissect the mechanism by which BACH2 induces its tumor
suppression. We identify that BACH2-mediated tumor suppression is largely dependent
on TP53. We also observe that the balance between BACH2 and BCL6 which is tightly
regulated during early B cell development is offset in leukemia. We find that PAX5 acts
upstream of Bach2 and modulates the levels of the latter during early B cell development.
Therefore, any mutation or translocation involving PAX5 in leukemia could prove to be
deleterious to a pre-B cell due to reduction in BACH2 levels.
All the conclusions drawn in this chapter univocally support the role of BACH2 as a
guardian against leukemogenesis at the pre-BCR checkpoint. Disruption of BACH2
would thus fail to eliminate deleterious B cells from the repertoire and eventually lead to
leukemia. Our studies also highlight the importance of BACH2 as a powerful predictor of
favorable clinical outcome in ALL patients, which would be particularly relevant for future
approaches of risk assessment.
6.5. Limitations and future perspectives
We propose that BACH2 levels would be relevant for risk assessment in childhood ALL
patients. Even though we show that BACH2 levels can be used as a predictor of clinical
outcome in ALL patients, we do not show sufficient multivariate analyses which
demonstrate the importance of BACH2 levels exclusively within high risk ALL patients.
One would thus need to carry out multivariate analyses for BACH2 levels using known
predictors of high risk like WBC count, IKZF and PAX5 deletions etc. Such analyses
would demonstrate whether BACH2 levels are useful in predicting relapse free survival
rates of high risk ALL patients.
Our current study proposes a mechanism of competition between BACH2 and BCL6 in
maintaining the balance between oncogene-induced apoptosis and leukemic
transformation. However, it would be useful to validate this competitive mechanism in
110
two ways- 1) Checking by co-immunoprecipitation whether BACH2 and BCL6 interact
directly with each other and, 2) Using luciferase assays to measure Arf/p53 promoter
activities after modulating relative levels of BACH2 and BCL6 inducibly. BACH2 and
BCL6 single locus ChIP studies are currently being carried out to measure the extent of
binding of these two proteins to promoters of common tumor suppressor genes.
Moreover, the study does not shed light on how such an understanding of the
mechanism of leukemogenesis could be used in a translational setting to treat ALL
patients. We thus propose that we could exploit the knowledge gained in this study to
selectively inhibit BCL6 using previously described inhibitors and thereby, shift the
balance from leukemic transformation to apoptosis. Such treatment studies need to be
verified using in vitro approaches and mouse models before extending them to the clinic.

111
Chapter 7. Discussion

Childhood acute lymphoblastic leukemia (ALL) arises from a pre-B cell clone that has
gone astray and is dividing indefinitely in the bone marrow. The process of conversion of
a normal pre-B cell to a leukemic pre-B cell is referred to as leukemogenesis. The
development of early B cells is a tightly controlled process with numerous checkpoints at
every stage that prevent its leukemic transformation (Chapter 1, Section 1.7). In this
thesis, we elucidate the two mechanisms by which a pre-B cell can bypass these
checkpoints to give rise to overt leukemia. Both the mechanisms of leukemogenesis
described in this work revolve around the pre-B cell receptor (Fraction C’/large pre-B
stage). The first mechanism involves the acquisition of AID and RAG induced-genetic
lesions due to a loss or completion of checkpoint signaling (here, IL7R signaling). The
second mechanism of leukemogenesis occurs when a deleterious pre-B cell fails to
undergo apoptosis (negative selection) as a result of deregulated checkpoint control
signaling (eg. loss of BACH2). The former mechanism can also lead to the latter by
mutating one of the key elements of the checkpoint control signaling pathway. For
example, loss of BACH2 function may occur as a result of being hypermutated by AID
(Jiang et al., 2012), which would block apoptosis at the critical checkpoints, and
ultimately lead to full-blown leukemia. As a summary, this thesis sheds light on two novel
failsafe barriers against leukemogenesis at the pre-BCR checkpoint, namely, IL7R and
BACH2.
In chapters 2 and 3, we lay down the importance of IL7R checkpoint at the Fraction C’-
Fraction D interface in safeguarding pre-B cells from acquisition of mutations.
Attenuation of IL7R signaling occurs during normal B cell development when cells
differentiate from Fraction C’ to D (Johnson et al., 2008). We show that loss of IL7R
signaling activates the notorious genes which have already been implicated in
112
leukemogenesis namely, AID deaminase and the RAG1/RAG2 recombinases (Tsai et al.,
2008).
The mechanism by which these two classes of genes cause genetic lesions is described
in chapters 1 and 2 of this thesis. We infer that IL7R signaling is not only crucial for
preventing leukemogenesis in mice but also in humans (Figure 3.5). Our findings identify
Fraction D as the subset of pre-B cells with increased genetic vulnerability as it displays
concomitant expression of both AID (Figure 3.2; Figure 3.3; Figure 3.4) and the RAGs
(Figure 3.11A). Following this, we conduct in-depth inquiry into the adaptors and
transcription factors which mediate the activation of these DNA damage- inducing genes.
On doing so, we identify STAT5 as a negative regulator (Figure 3.6; Johnson et al.,
2012) and the FoxO factors (Figure 3.7; Figure 3.11; Kuo et al., 2009) as the
transcriptional repressor and activators of Aid and Rag enzymes at Fraction D,
respectively.
We also show experimentally that the process of acquisition of mutations at Fraction D is
not countered by protective mechanisms that are normally triggered in a germinal center
upon infection. For example, miR155 which is a negative regulator of AID and a buffer
against DNA damage inflicted by AID (Teng et al., 2008), is not upregulated at Fraction
D (Figure 3.10). Our results demonstrate that such an absence of safeguard
mechanisms like the IL7R checkpoint and miR155 upregulation, make a Fraction D pre-
B cell more susceptible to external cues like inflammation or infection (Figure 3.8; Figure
3.9).
We observe that our findings are in agreement with the “Delayed Infections Hypothesis”
proposed by Dr. Mel Greaves and Dr. Joseph Wiemels and clarify the mechanism
behind TEL-AML1-driven childhood leukemia. We find that AID and RAGs are required
to convert a TEL-AML1
+
Fraction D pre-B cell to a leukemic cell upon repeated infection
113
or inflammation (Figure 3.13). We also show that the leukemia derived as a result of this
is clonal in origin (Figure 3.14).
In conclusion, all our findings in the first half of the thesis give one example of how
natural mechanisms existing in a Fraction C’ pre-B cell protect against leukemia by
preventing genetic alterations.
In chapters 4, 5 and 6 of this thesis, we focus on the second mechanism of
leukemogenesis which results from the inhibition of pre-BCR-induced negative selection.
The pre-BCR checkpoint control is explained in chapter 1. The negative selection
process at the pre-BCR checkpoint is required for elimination of B cell clones with non-
functional IgH rearrangements. Our findings identify the BTB and leucine zipper
containing B-lymphoid protein BACH2 as the key inducer of negative selection (chapter
5).
Using a knockout mouse model of Bach2, we show that loss of Bach2 completely
abrogates negative selection at the pre-BCR checkpoint (Figure 5.1). We also show that
this defect can be rescued by inducible overexpression of Bach2 in Bach2-deficient
murine pre-B cells (Figure 5.2). We then characterize the molecules involved in the
negative selection signaling pathway and identify PAX5 as the key transactivator of
Bach2 and ARF and p53 as its effectors (Figure 5.3). In addition, we find that BACH2
increases the level and activity of the RAG enzymes (Figure 5.4; Figure 5.5; Figure 5.6).
We propose that by modulating the RAGs, BACH2 increases the probability of a pre-B
cell to successfully rearrange the IgH. Another theory is that BACH2-induced Rag1 and
Rag2 expression allows the cell to acquire enough DNA damage so that negative
selection (apoptosis) can be easily triggered. Further investigations are required to test if
the latter theory is true.
Our next goal was to identify the key signaling molecules that maintain the balance
between negative selection and survival at the pre-BCR checkpoint. Earlier studies from
114
our lab showed that BCL6 is essential for survival at the pre-BCR checkpoint (Duy et al.,
2010). We therefore hypothesized that relative levels of BACH2 and BCL6 would decide
between negative selection and survival at this checkpoint. Such a balance would also
decide between two other similar opposing processes, namely, leukemogenesis and
apoptosis.
Based on our studies on early B cell development which highlight the importance of
BACH2 in negative selection, we propose that it is a putative tumor suppressor in ALL
for two reasons: 1) Bach2 is required for induction of apoptosis at the pre-BCR
checkpoint and, 2) Bach2 is activated by PAX5, and the latter has been characterized as
a tumor suppressor in pre-B ALL (Mullighan et al., 2007, Medvedovic et al., 2011).
We begin our study of BACH2 in ALL by analyzing a large cohort of childhood ALL
patients (COG P9906) to check if there is any relationship between BACH2 levels in
these patients and their disease status (eg. RFS, MRD, risk etc). We identify an inverse
relationship between BACH2 levels and the severity of the disease in each study (Figure
6.1-6.5). All the patient studies point to high BACH2 levels as a predictor of favorable
clinical outcome. These studies and the ones in chapter 5 univocally support a tumor
suppressive role for BACH2.
Using a Bach2-deficient mouse model, we identify its requirement in blocking or slowing
down the process of leukemic transformation by oncogenes like BCR-ABL1 (Figure 6.6),
c-MYC (Figure 6.7-6.9, Figure 6.11) and STAT5
CA
(Figure 6.10). We also show that loss
of Bach2 occurs during progressive leukemic transformation (Figure 6.12). These
studies provide concrete evidence for BACH2 as a failsafe barrier against
leukemogenesis at the pre-BCR checkpoint.
Following this, we analyze the cell signaling pathway of Bach2-induced leukemia
suppression. As in the case of early B cell development, we identify PAX5 as the
upstream activator of BACH2 (Figure 6.23). We demonstrate that BACH2 induces
115
apoptosis through the classical Arf-Mdm2-p53 tumor suppressor pathway (Figure C.3,
Figure 5.3, Figure 6.13, Figure 6.14A) and that p53 is an absolute requirement for this
process of programmed cell death (Figure 6.21-6.22). We show that relative levels of
BACH2 and BCL6 determine whether leukemogenesis or apoptosis is triggered after
oncogenic insult (Figure 6.15-6.20). These studies shed light on how a slight
deregulation in checkpoint control signaling can tilt the balance from normal B cell
development to leukemic transformation.
Genetic instability and resistance to apoptosis are two major hallmarks of all cancers
(Hanahan and Weinberg, 2000). In this thesis, we examine how a large pre-B cell
acquires both the above mentioned hallmarks and progresses towards full-blown
leukemia. So far, this is the first study highlighting signaling pathways at the pre-BCR
checkpoint that are crucial to control genetic instability, and induce programmed cell
death. Using one example from the context of B cell development, for each of the two
hallmarks, we elucidate the exact mechanism of leukemogenesis (Figure 7.1). Such an
understanding of the mechanism of leukemogenesis is essential for optimizing treatment
strategies for ALL patients.
Figure 7.1. IL7R signaling and BACH2 represent two failsafe barriers against
leukemogenesis at the pre-BCR checkpoint.
Leukemogenesis can occur as a result of deregulated cell signaling mechanisms at the
pre-BCR checkpoint. Shown below are two mechanisms of this deregulation, namely,
acquisition of genetic alterations due to inflammatory cues, and, survival of a leukemic
pre-B cell as opposed to its apoptotic elimination (negative selection). IL7R signaling is
the guardian against pre-B cell leukemogenesis resulting from the former mechanism.
BACH2 and PAX5 are the guardians against leukemia that develops through the latter
mechanism. Boldface indicates high levels of a particular protein. All components
indicated in red favor leukemogenesis and the ones indicated in green suppress it. Loss
of any one of the checkpoints indicated in green favors leukemia initiation.

116
Figure 7.1. continued.

p53
117
Bibliography

Agrawal, A., Eastman, Q.M., and Schatz, D.G. (1998). Transposition mediated by RAG1
and RAG2 and its implications for the evolution of the immune system. Nature 394, 744–
751.

Aspland, S.E., Bendall, H.H., and Murre, C. (2001). The role of E2A-PBX1 in
leukemogenesis. Oncogene 20, 5708–5717.

Bertrand, F.E., Vogtenhuber, C., Shah, N., and LeBien, T.W. (2001). Pro-B-cell to pre-B-
cell development in B-lineage acute lymphoblastic leukemia expressing the MLL/AF4
fusion protein. Blood 98, 3398–3405.

Blier, P.R., Griffith, A.J., Craft, J., and Hardin, J.A. (1993). Binding of Ku protein to DNA.
Measurement of affinity for ends and demonstration of binding to nicks. J. Biol. Chem.
268, 7594–7601.

Borkhardt, A., Cazzaniga, G., Viehmann, S., Valsecchi, M.G., Ludwig, W.D., Burci, L.,
Mangioni, S., Schrappe, M., Riehm, H., Lampert, F., et al. (1997). Incidence and clinical
relevance of TEL/AML1 fusion genes in children with acute lymphoblastic leukemia
enrolled in the German and Italian multicenter therapy trials. Associazione Italiana
Ematologia Oncologia Pediatrica and the Berlin-Frankfurt-Münster Study Group. Blood
90, 571–577.

Busslinger, M., Nutt, S.L., and Rolink, A.G. (2000). Lineage commitment in
lymphopoiesis. Curr. Opin. Immunol. 12, 151–158.

Casolari, D.A., Makri, M., Yoshida, C., Muto, A., Igarashi, K., and Melo, J.V. (2012).
Transcriptional suppression of BACH2 by the Bcr-Abl oncoprotein is mediated by PAX5.
Leukemia.

Chiu, C.W., Dalton, M., Ishiai, M., Kurosaki, T., and Chan, A.C. (2002). BLNK: molecular
scaffolding through “cis”-mediated organization of signaling proteins. EMBO J. 21, 6461–
6472.

Cobaleda, C., Schebesta, A., Delogu, A., and Busslinger, M. (2007). Pax5: the guardian
of B cell identity and function. Nat. Immunol. 8, 463–470.

Crouch, E.E., Li, Z., Takizawa, M., Fichtner-Feigl, S., Gourzi, P., Montaño, C.,
Feigenbaum, L., Wilson, P., Janz, S., Papavasiliou, F.N., et al. (2007). Regulation of AID
expression in the immune response. J. Exp. Med. 204, 1145–1156.

Cui, Y., Riedlinger, G., Miyoshi, K., Tang, W., Li, C., Deng, C.-X., Robinson, G.W., and
Hennighausen, L. (2004). Inactivation of Stat5 in mouse mammary epithelium during
pregnancy reveals distinct functions in cell proliferation, survival, and differentiation. Mol.
Cell. Biol. 24, 8037–8047.

Deane, J.A., and Fruman, D.A. (2004). Phosphoinositide 3-kinase: diverse roles in
immune cell activation. Annu. Rev. Immunol. 22, 563–598.

118
Deininger, M.W., Goldman, J.M., and Melo, J.V. (2000). The molecular biology of
chronic myeloid leukemia. Blood 96, 3343–3356.

DeKoter, R.P., and Singh, H. (2000). Regulation of B lymphocyte and macrophage
development by graded expression of PU.1. Science 288, 1439–1441.

Duy, C., Hurtz, C., Shojaee, S., Cerchietti, L., Geng, H., Swaminathan, S., Klemm, L.,
Kweon, S., Nahar, R., Braig, M., et al. (2011). BCL6 enables Ph+ acute lymphoblastic
leukaemia cells to survive BCR-ABL1 kinase inhibition. Nature 473, 384–388.

Duy, C., Yu, J.J., Nahar, R., Swaminathan, S., Kweon, S.-M., Polo, J.M., Valls, E.,
Klemm, L., Shojaee, S., Cerchietti, L., et al. (2010). BCL6 is critical for the development
of a diverse primary B cell repertoire. J. Exp. Med. 207, 1209–1221.

Eischen, C.M., Weber, J.D., Roussel, M.F., Sherr, C.J., and Cleveland, J.L. (1999).
Disruption of the ARF-Mdm2-p53 tumor suppressor pathway in Myc-induced
lymphomagenesis. Genes & Development 13, 2658–2669.

Faderl, S., Kantarjian, H.M., Thomas, D.A., Cortes, J., Giles, F., Pierce, S., Albitar, M.,
and Estrov, Z. (2000). Outcome of Philadelphia chromosome-positive adult acute
lymphoblastic leukemia. Leuk. Lymphoma 36, 263–273.

Feldhahn, N., Henke, N., Melchior, K., Duy, C., Soh, B.N., Klein, F., Von Levetzow, G.,
Giebel, B., Li, A., Hofmann, W.-K., et al. (2007). Activation-induced cytidine deaminase
acts as a mutator in BCR-ABL1-transformed acute lymphoblastic leukemia cells. J. Exp.
Med. 204, 1157–1166.

Ford, A.M., Bennett, C.A., Price, C.M., Bruin, M.C., Van Wering, E.R., and Greaves, M.
(1998). Fetal origins of the TEL-AML1 fusion gene in identical twins with leukemia. Proc.
Natl. Acad. Sci. U.S.A. 95, 4584–4588.

Ford, A.M., Palmi, C., Bueno, C., Hong, D., Cardus, P., Knight, D., Cazzaniga, G., Enver,
T., and Greaves, M. (2009). The TEL-AML1 leukemia fusion gene dysregulates the
TGF-beta pathway in early B lineage progenitor cells. J. Clin. Invest. 119, 826–836.

Gale, K.B., Ford, A.M., Repp, R., Borkhardt, A., Keller, C., Eden, O.B., and Greaves,
M.F. (1997). Backtracking leukemia to birth: identification of clonotypic gene fusion
sequences in neonatal blood spots. Proc. Natl. Acad. Sci. U.S.A. 94, 13950–13954.

Galy, A., Travis, M., Cen, D., and Chen, B. (1995). Human T, B, natural killer, and
dendritic cells arise from a common bone marrow progenitor cell subset. Immunity 3,
459–473.

Gellert, M. (2002). V(D)J recombination: RAG proteins, repair factors, and regulation.
Annu. Rev. Biochem. 71, 101–132.

Georgopoulos, K., Bigby, M., Wang, J.H., Molnar, A., Wu, P., Winandy, S., and Sharpe,
A. (1994). The Ikaros gene is required for the development of all lymphoid lineages. Cell
79, 143–156.

119
Goetz, C.A., Harmon, I.R., O’Neil, J.J., Burchill, M.A., and Farrar, M.A. (2004). STAT5
Activation Underlies IL7 Receptor-Dependent B Cell Development. J Immunol 172,
4770–4778.

Grawunder, U., Zimmer, D., Fugmann, S., Schwarz, K., and Lieber, M.R. (1998). DNA
ligase IV is essential for V(D)J recombination and DNA double-strand break repair in
human precursor lymphocytes. Mol. Cell 2, 477–484.

Greaves, M., and Maley, C.C. (2012). Clonal evolution in cancer. Nature 481, 306–313.

Greaves, M.F., Maia, A.T., Wiemels, J.L., and Ford, A.M. (2003). Leukemia in twins:
lessons in natural history. Blood 102, 2321–2333.

Greaves, M.F., and Wiemels, J. (2003). Origins of chromosome translocations in
childhood leukaemia. Nat. Rev. Cancer 3, 639–649.

Green, M., Gandhi, M.K., Camilleri, E., Marlton, P., Lea, R., and Griffiths, L. (2009). High
levels of BACH2 associated with lower levels of BCL2 transcript abundance in
t(14;18)(q21;q34) translocation positive non-Hodgkin’s lymphoma. Leuk. Res. 33, 731–
734.

Gruber, T.A., Chang, M.S., Sposto, R., and Müschen, M. (2010). Activation-induced
cytidine deaminase accelerates clonal evolution in BCR-ABL1-driven B-cell lineage
acute lymphoblastic leukemia. Cancer Res. 70, 7411–7420.

Gu, J., Lu, H., Tippin, B., Shimazaki, N., Goodman, M.F., and Lieber, M.R. (2007a).
XRCC4:DNA ligase IV can ligate incompatible DNA ends and can ligate across gaps.
EMBO J. 26, 1010–1023.

Gu, J., Lu, H., Tsai, A.G., Schwarz, K., and Lieber, M.R. (2007b). Single-stranded DNA
ligation and XLF-stimulated incompatible DNA end ligation by the XRCC4-DNA ligase IV
complex: influence of terminal DNA sequence. Nucleic Acids Res. 35, 5755–5762.

Guo, B., Kato, R.M., Garcia-Lloret, M., Wahl, M.I., and Rawlings, D.J. (2000).
Engagement of the human pre-B cell receptor generates a lipid raft-dependent calcium
signaling complex. Immunity 13, 243–253.

Han, J.-H., Akira, S., Calame, K., Beutler, B., Selsing, E., and Imanishi-Kari, T. (2007).
Class switch recombination and somatic hypermutation in early mouse B cells are
mediated by B cell and Toll-like receptors. Immunity 27, 64–75.

Hanahan, D., and Weinberg, R.A. (2000). The hallmarks of cancer. Cell 100, 57–70.

Hardy, R.R., and Hayakawa, K. (2001). B cell development pathways. Annu. Rev.
Immunol. 19, 595–621.

Harvey, R.C., Mullighan, C.G., Wang, X., Dobbin, K.K., Davidson, G.S., Bedrick, E.J.,
Chen, I.-M., Atlas, S.R., Kang, H., Ar, K., et al. (2010). Identification of novel cluster
groups in pediatric high-risk B-precursor acute lymphoblastic leukemia with gene
expression profiling: correlation with genome-wide DNA copy number alterations, clinical
characteristics, and outcome. Blood 116, 4874–4884.
120

Hashimoto, S., Iwamatsu, A., Ishiai, M., Okawa, K., Yamadori, T., Matsushita, M., Baba,
Y., Kishimoto, T., Kurosaki, T., and Tsukada, S. (1999). Identification of the SH2 domain
binding protein of Bruton’s tyrosine kinase as BLNK--functional significance of Btk-SH2
domain in B-cell antigen receptor-coupled calcium signaling. Blood 94, 2357–2364.

Heerema, N.A., Sather, H.N., Ge, J., Arthur, D.C., Hilden, J.M., Trigg, M.E., and
Reaman, G.H. (1999). Cytogenetic studies of infant acute lymphoblastic leukemia: poor
prognosis of infants with t(4;11) - a report of the Children’s Cancer Group. Leukemia 13,
679–686.

Heisterkamp, N., Jenster, G., Ten Hoeve, J., Zovich, D., Pattengale, P.K., and Groffen, J.
(1990). Acute leukaemia in bcr/abl transgenic mice. Nature 344, 251–253.

Hendriks, R.W., and Middendorp, S. (2004). The pre-BCR checkpoint as a cell-
autonomous proliferation switch. Trends in Immunology 25, 249–256.

Herzog, S., Hug, E., Meixlsperger, S., Paik, J.-H., DePinho, R.A., Reth, M., and Jumaa,
H. (2008). SLP-65 regulates immunoglobulin light chain gene recombination through the
PI(3)K-PKB-Foxo pathway. Nat. Immunol. 9, 623–631.

Herzog, S., Reth, M., and Jumaa, H. (2009). Regulation of B-cell proliferation and
differentiation by pre-B-cell receptor signalling. Nat. Rev. Immunol. 9, 195–205.

Hiom, K., Melek, M., and Gellert, M. (1998). DNA transposition by the RAG1 and RAG2
proteins: a possible source of oncogenic translocations. Cell 94, 463–470.

Hogan, L.E., Meyer, J.A., Yang, J., Wang, J., Wong, N., Yang, W., Condos, G., Hunger,
S.P., Raetz, E., Saffery, R., et al. (2011). Integrated genomic analysis of relapsed
childhood acute lymphoblastic leukemia reveals therapeutic strategies. Blood 118,
5218–5226.

Hoshino, H., Kobayashi, A., Yoshida, M., Kudo, N., Oyake, T., Motohashi, H., Hayashi,
N., Yamamoto, M., and Igarashi, K. (2000). Oxidative stress abolishes leptomycin B-
sensitive nuclear export of transcription repressor Bach2 that counteracts activation of
Maf recognition element. J. Biol. Chem. 275, 15370–15376.

Hunger, S.P., Galili, N., Carroll, A.J., Crist, W.M., Link, M.P., and Cleary, M.L. (1991).
The t(1;19)(q23;p13) results in consistent fusion of E2A and PBX1 coding sequences in
acute lymphoblastic leukemias. Blood 77, 687–693.

Igarashi, H., Gregory, S.C., Yokota, T., Sakaguchi, N., and Kincade, P.W. (2002).
Transcription from the RAG1 locus marks the earliest lymphocyte progenitors in bone
marrow. Immunity 17, 117–130.

Ishiai, M., Kurosaki, M., Pappu, R., Okawa, K., Ronko, I., Fu, C., Shibata, M., Iwamatsu,
A., Chan, A.C., and Kurosaki, T. (1999). BLNK required for coupling Syk to PLC gamma
2 and Rac1-JNK in B cells. Immunity 10, 117–125.

121
Jiang, Y., Soong, T.D., Wang, L., Melnick, A.M., and Elemento, O. (2012). Genome-wide
detection of genes targeted by non-Ig somatic hypermutation in lymphoma. PLoS ONE 7,
e40332.

Johnson, K., Hashimshony, T., Sawai, C.M., Pongubala, J.M.R., Skok, J.A., Aifantis, I.,
and Singh, H. (2008). Regulation of immunoglobulin light-chain recombination by the
transcription factor IRF-4 and the attenuation of interleukin-7 signaling. Immunity 28,
335–345.

Jumaa, H., Wollscheid, B., Mitterer, M., Wienands, J., Reth, M., and Nielsen, P.J. (1999).
Abnormal development and function of B lymphocytes in mice deficient for the signaling
adaptor protein SLP-65. Immunity 11, 547–554.

Kamio, T., Toki, T., Kanezaki, R., Sasaki, S., Tandai, S., Terui, K., Ikebe, D., Igarashi, K.,
and Ito, E. (2003). B-cell-specific transcription factor BACH2 modifies the cytotoxic
effects of anticancer drugs. Blood 102, 3317–3322.

Kanie, T., Abe, A., Matsuda, T., Kuno, Y., Towatari, M., Yamamoto, T., Saito, H., Emi, N.,
and Naoe, T. (2004). TEL-Syk fusion constitutively activates PI3-K/Akt, MAPK and
JAK2-independent STAT5 signal pathways. Leukemia 18, 548–555.

Karasuyama, H., Rolink, A., Shinkai, Y., Young, F., Alt, F.W., and Melchers, F. (1994).
The expression of Vpre-B/lambda 5 surrogate light chain in early bone marrow precursor
B cells of normal and B cell-deficient mutant mice. Cell 77, 133–143.

Kawamata, N., Pennella, M.A., Woo, J.L., Berk, A.J., and Koeffler, H.P. (2012).
Dominant-negative mechanism of leukemogenic PAX5 fusions. Oncogene 31, 966–977.

Kikuchi, K., Kasai, H., Watanabe, A., Lai, A.Y., and Kondo, M. (2008). IL-7 specifies B
cell fate at the common lymphoid progenitor to pre-proB transition stage by maintaining
early B cell factor expression. J. Immunol. 181, 383–392.

Kim-Rouille, M.H., MacGregor, A., Wiedemann, L.M., Greaves, M.F., and Navarrete, C.
(1999). MLL-AF4 gene fusions in normal newborns. Blood 93, 1107–1108.

Klemm, L., Duy, C., Iacobucci, I., Kuchen, S., Von Levetzow, G., Feldhahn, N., Henke,
N., Li, Z., Hoffmann, T.K., Kim, Y., et al. (2009). The B cell mutator AID promotes B
lymphoid blast crisis and drug resistance in chronic myeloid leukemia. Cancer Cell 16,
232–245.

Kobayashi, A., Yamagiwa, H., Hoshino, H., Muto, A., Sato, K., Morita, M., Hayashi, N.,
Yamamoto, M., and Igarashi, K. (2000). A combinatorial code for gene expression
generated by transcription factor Bach2 and MAZR (MAZ-related factor) through the
BTB/POZ domain. Mol. Cell. Biol. 20, 1733–1746.

Kuo, T.C., and Schlissel, M.S. (2009). Mechanisms controlling expression of the RAG
locus during lymphocyte development. Curr. Opin. Immunol. 21, 173–178.

Lesche, R., Groszer, M., Gao, J., Wang, Y., Messing, A., Sun, H., Liu, X., and Wu, H.
(2002). Cre/loxP-mediated inactivation of the murine Pten tumor suppressor gene.
Genesis 32, 148–149.
122

Lewis, S., Rosenberg, N., Alt, F., and Baltimore, D. (1982). Continuing kappa-gene
rearrangement in a cell line transformed by Abelson murine leukemia virus. Cell 30,
807–816.

Li, Y.S., Wasserman, R., Hayakawa, K., and Hardy, R.R. (1996). Identification of the
earliest B lineage stage in mouse bone marrow. Immunity 5, 527–535.

Lieber, M.R. (2010). The mechanism of double-strand DNA break repair by the
nonhomologous DNA end-joining pathway. Annu. Rev. Biochem. 79, 181–211.

Lieber, M.R., Ma, Y., Pannicke, U., and Schwarz, K. (2004). The mechanism of
vertebrate nonhomologous DNA end joining and its role in V(D)J recombination. DNA
Repair (Amst.) 3, 817–826.

Liu, J., Sørensen, A.B., Wang, B., Wabl, M., Nielsen, A.L., and Pedersen, F.S. (2009).
Identification of novel Bach2 transcripts and protein isoforms through tagging analysis of
retroviral integrations in B-cell lymphomas. BMC Mol. Biol. 10, 2.

Liu, M., Duke, J.L., Richter, D.J., Vinuesa, C.G., Goodnow, C.C., Kleinstein, S.H., and
Schatz, D.G. (2008). Two levels of protection for the B cell genome during somatic
hypermutation. Nature 451, 841–845.

Look, A.T. (1997). Oncogenic transcription factors in the human acute leukemias.
Science 278, 1059–1064.

Ma, X., Buffler, P.A., Selvin, S., Matthay, K.K., Wiencke, J.K., Wiemels, J.L., and
Reynolds, P. (2002a). Daycare attendance and risk of childhood acute lymphoblastic
leukaemia. Br. J. Cancer 86, 1419–1424.

Ma, Y., Pannicke, U., Schwarz, K., and Lieber, M.R. (2002b). Hairpin opening and
overhang processing by an Artemis/DNA-dependent protein kinase complex in
nonhomologous end joining and V(D)J recombination. Cell 108, 781–794.

Matsuo, Y., and Drexler, H.G. (1998). Establishment and characterization of human B
cell precursor-leukemia cell lines. Leuk. Res. 22, 567–579.

Medvedovic, J., Ebert, A., Tagoh, H., and Busslinger, M. (2011). Pax5: a master
regulator of B cell development and leukemogenesis. Adv. Immunol. 111, 179–206.

Mombaerts, P., Iacomini, J., Johnson, R.S., Herrup, K., Tonegawa, S., and Papaioannou,
V.E. (1992). RAG-1-deficient mice have no mature B and T lymphocytes. Cell 68, 869–
877.

Mostoslavsky, R., Alt, F.W., and Rajewsky, K. (2004). The lingering enigma of the allelic
exclusion mechanism. Cell 118, 539–544.

Mullighan, C.G., Goorha, S., Radtke, I., Miller, C.B., Coustan-Smith, E., Dalton, J.D.,
Girtman, K., Mathew, S., Ma, J., Pounds, S.B., et al. (2007). Genome-wide analysis of
genetic alterations in acute lymphoblastic leukaemia. Nature 446, 758–764.

123
Muramatsu, M., Kinoshita, K., Fagarasan, S., Yamada, S., Shinkai, Y., and Honjo, T.
(2000). Class switch recombination and hypermutation require activation-induced
cytidine deaminase (AID), a potential RNA editing enzyme. Cell 102, 553–563.

Muramatsu, M., Sankaranand, V.S., Anant, S., Sugai, M., Kinoshita, K., Davidson, N.O.,
and Honjo, T. (1999). Specific expression of activation-induced cytidine deaminase (AID),
a novel member of the RNA-editing deaminase family in germinal center B cells. J. Biol.
Chem. 274, 18470–18476.

Muto, A., Hoshino, H., Madisen, L., Yanai, N., Obinata, M., Karasuyama, H., Hayashi, N.,
Nakauchi, H., Yamamoto, M., Groudine, M., et al. (1998). Identification of Bach2 as a B-
cell-specific partner for small maf proteins that negatively regulate the immunoglobulin
heavy chain gene 3’ enhancer. EMBO J. 17, 5734–5743.

Muto, A., Ochiai, K., Kimura, Y., Itoh-Nakadai, A., Calame, K.L., Ikebe, D., Tashiro, S.,
and Igarashi, K. (2010). Bach2 represses plasma cell gene regulatory network in B cells
to promote antibody class switch. EMBO J. 29, 4048–4061.

Muto, A., Tashiro, S., Nakajima, O., Hoshino, H., Takahashi, S., Sakoda, E., Ikebe, D.,
Yamamoto, M., and Igarashi, K. (2004). The transcriptional programme of antibody class
switching involves the repressor Bach2. Nature 429, 566–571.

Muto, A., Tashiro, S., Tsuchiya, H., Kume, A., Kanno, M., Ito, E., Yamamoto, M., and
Igarashi, K. (2002). Activation of Maf/AP-1 repressor Bach2 by oxidative stress promotes
apoptosis and its interaction with promyelocytic leukemia nuclear bodies. J. Biol. Chem.
277, 20724–20733.

Nosaka, T., Kawashima, T., Misawa, K., Ikuta, K., Mui, A.L., and Kitamura, T. (1999).
STAT5 as a molecular regulator of proliferation, differentiation and apoptosis in
hematopoietic cells. EMBO J. 18, 4754–4765.

NOWELL, P.C., and HUNGERFORD, D.A. (1960). Chromosome studies on normal and
leukemic human leukocytes. J. Natl. Cancer Inst. 25, 85–109.

Nutt, S.L., Heavey, B., Rolink, A.G., and Busslinger, M. (1999). Commitment to the B-
lymphoid lineage depends on the transcription factor Pax5. Nature 401, 556–562.

Ono, A., Kono, K., Ikebe, D., Muto, A., Sun, J., Kobayashi, M., Ueda, K., Melo, J.V.,
Igarashi, K., and Tashiro, S. (2007). Nuclear positioning of the BACH2 gene in BCR-ABL
positive leukemic cells. Genes Chromosomes Cancer 46, 67–74.

Oyake, T., Itoh, K., Motohashi, H., Hayashi, N., Hoshino, H., Nishizawa, M., Yamamoto,
M., and Igarashi, K. (1996). Bach proteins belong to a novel family of BTB-basic leucine
zipper transcription factors that interact with MafK and regulate transcription through the
NF-E2 site. Mol. Cell. Biol. 16, 6083–6095.

Padilla-Nash, H.M., Barenboim-Stapleton, L., Difilippantonio, M.J., and Ried, T. (2006).
Spectral karyotyping analysis of human and mouse chromosomes. Nat Protoc 1, 3129–
3142.

124
Perrillat, F., Clavel, J., Auclerc, M.F., Baruchel, A., Leverger, G., Nelken, B., Philippe, N.,
Schaison, G., Sommelet, D., Vilmer, E., et al. (2002). Day-care, early common infections
and childhood acute leukaemia: a multicentre French case-control study. Br. J. Cancer
86, 1064–1069.

Plaisier, S.B., Taschereau, R., Wong, J.A., and Graeber, T.G. (2010). Rank-rank
hypergeometric overlap: identification of statistically significant overlap between gene-
expression signatures. Nucleic Acids Res. 38, e169.

Plum, J., De Smedt, M., Leclercq, G., Verhasselt, B., and Vandekerckhove, B. (1996).
Interleukin-7 is a critical growth factor in early human T-cell development. Blood 88,
4239–4245.

Post, S.M., Quintás-Cardama, A., Terzian, T., Smith, C., Eischen, C.M., and Lozano, G.
(2010). p53-dependent senescence delays Emu-myc-induced B-cell lymphomagenesis.
Oncogene 29, 1260–1269.

Pui, C.-H., Relling, M.V., and Downing, J.R. (2004). Acute lymphoblastic leukemia. N.
Engl. J. Med. 350, 1535–1548.

Raimondi, S. C. (1999). Childhood Leukemias (ed. Pui, C.-H.) 168-196 (Cambridge Univ.
Press, Cambridge).

Rasmussen, M.H., Ballarín-González, B., Liu, J., Lassen, L.B., Füchtbauer, A.,
Füchtbauer, E.-M., Nielsen, A.L., and Pedersen, F.S. (2010). Antisense transcription in
gammaretroviruses as a mechanism of insertional activation of host genes. J. Virol. 84,
3780–3788.

Raynaud, S., Cave, H., Baens, M., Bastard, C., Cacheux, V., Grosgeorge, J., Guidal-
Giroux, C., Guo, C., Vilmer, E., Marynen, P., et al. (1996). The 12;21 translocation
involving TEL and deletion of the other TEL allele: two frequently associated alterations
found in childhood acute lymphoblastic leukemia. Blood 87, 2891–2899.

Reynaud, D., Demarco, I.A., Reddy, K.L., Schjerven, H., Bertolino, E., Chen, Z., Smale,
S.T., Winandy, S., and Singh, H. (2008). Regulation of B cell fate commitment and
immunoglobulin heavy-chain gene rearrangements by Ikaros. Nat. Immunol. 9, 927–936.

Rolink, A.G., Ten Boekel, E., Yamagami, T., Ceredig, R., Andersson, J., and Melchers, F.
(1999). B cell development in the mouse from early progenitors to mature B cells.
Immunol. Lett. 68, 89–93.

Rolink, A.G., Schaniel, C., Busslinger, M., Nutt, S.L., and Melchers, F. (2000a). Fidelity
and infidelity in commitment to B-lymphocyte lineage development. Immunol. Rev. 175,
104–111.

Rolink, A.G., Winkler, T., Melchers, F., and Andersson, J. (2000b). Precursor B cell
receptor-dependent B cell proliferation and differentiation does not require the bone
marrow or fetal liver environment. J. Exp. Med. 191, 23–32.

125
Romana, S.P., Mauchauffé, M., Le Coniat, M., Chumakov, I., Le Paslier, D., Berger, R.,
and Bernard, O.A. (1995a). The t(12;21) of acute lymphoblastic leukemia results in a tel-
AML1 gene fusion. Blood 85, 3662–3670.

Romana, S.P., Poirel, H., Leconiat, M., Flexor, M.A., Mauchauffé, M., Jonveaux, P.,
Macintyre, E.A., Berger, R., and Bernard, O.A. (1995b). High frequency of t(12;21) in
childhood B-lineage acute lymphoblastic leukemia. Blood 86, 4263–4269.

Ross, M.E., Zhou, X., Song, G., Shurtleff, S.A., Girtman, K., Williams, W.K., Liu, H.-C.,
Mahfouz, R., Raimondi, S.C., Lenny, N., et al. (2003). Classification of pediatric acute
lymphoblastic leukemia by gene expression profiling. Blood 102, 2951–2959.

Rouault, J.P., Falette, N., Guéhenneux, F., Guillot, C., Rimokh, R., Wang, Q., Berthet, C.,
Moyret-Lalle, C., Savatier, P., Pain, B., et al. (1996). Identification of BTG2, an
antiproliferative p53-dependent component of the DNA damage cellular response
pathway. Nat. Genet. 14, 482–486.

Sakane-Ishikawa, E., Nakatsuka, S.-I., Tomita, Y., Fujita, S., Nakamichi, I., Takakuwa,
T., Sugiyama, H., Fukuhara, S., Hino, M., Kanamaru, A., et al. (2005). Prognostic
significance of BACH2 expression in diffuse large B-cell lymphoma: a study of the Osaka
Lymphoma Study Group. J. Clin. Oncol. 23, 8012–8017.

Sasaki, S., Ito, E., Toki, T., Maekawa, T., Kanezaki, R., Umenai, T., Muto, A., Nagai, H.,
Kinoshita, T., Yamamoto, M., et al. (2000). Cloning and expression of human B cell-
specific transcription factor BACH2 mapped to chromosome 6q15. Oncogene 19, 3739–
3749.

Sattler, M., and Salgia, R. (1997). Activation of hematopoietic growth factor signal
transduction pathways by the human oncogene BCR/ABL. Cytokine Growth Factor Rev.
8, 63–79.

Schatz, D.G., Oettinger, M.A., and Baltimore, D. (1989). The V(D)J recombination
activating gene, RAG-1. Cell 59, 1035–1048.

Shimizu, T., Mundt, C., Licence, S., Melchers, F., and Mårtensson, I.-L. (2002).
VpreB1/VpreB2/lambda 5 triple-deficient mice show impaired B cell development but
functional allelic exclusion of the IgH locus. J. Immunol. 168, 6286–6293.

Shurtleff, S.A., Buijs, A., Behm, F.G., Rubnitz, J.E., Raimondi, S.C., Hancock, M.L.,
Chan, G.C., Pui, C.H., Grosveld, G., and Downing, J.R. (1995). TEL/AML1 fusion
resulting from a cryptic t(12;21) is the most common genetic lesion in pediatric ALL and
defines a subgroup of patients with an excellent prognosis. Leukemia 9, 1985–1989.

Staszewski, O., Baker, R.E., Ucher, A.J., Martier, R., Stavnezer, J., and Guikema, J.E.J.
(2011). Activation-induced cytidine deaminase induces reproducible DNA breaks at
many non-Ig Loci in activated B cells. Mol. Cell 41, 232–242.

Streubel, B., Vinatzer, U., Willheim, M., Raderer, M., and Chott, A. (2006). Novel
t(5;9)(q33;q22) fuses ITK to SYK in unspecified peripheral T-cell lymphoma. Leukemia
20, 313–318.

126
Su, Y.W., Zhang, Y., Schweikert, J., Koretzky, G.A., Reth, M., and Wienands, J. (1999).
Interaction of SLP adaptors with the SH2 domain of Tec family kinases. Eur. J. Immunol.
29, 3702–3711.

Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A.,
Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., et al. (2005). Gene set
enrichment analysis: a knowledge-based approach for interpreting genome-wide
expression profiles. Proc. Natl. Acad. Sci. U.S.A. 102, 15545–15550.

Sudo, T., Nishikawa, S., Ohno, N., Akiyama, N., Tamakoshi, M., Yoshida, H., and
Nishikawa, S. (1993). Expression and function of the interleukin 7 receptor in murine
lymphocytes. Proc. Natl. Acad. Sci. U.S.A. 90, 9125–9129.

Ta, V.B.T., De Bruijn, M.J.W., Ter Brugge, P.J., Van Hamburg, J.P., Diepstraten, H.J.A.,
Van Loo, P.F., Kersseboom, R., and Hendriks, R.W. (2010). Malignant transformation of
Slp65-deficient pre-B cells involves disruption of the Arf-Mdm2-p53 tumor suppressor
pathway. Blood 115, 1385–1393.

Takakuwa, T., Luo, W.-J., Ham, M.F., Sakane-Ishikawa, F., Wada, N., and Aozasa, K.
(2004). Integration of Epstein-Barr virus into chromosome 6q15 of Burkitt lymphoma cell
line (Raji) induces loss of BACH2 expression. Am. J. Pathol. 164, 967–974.

Teng, G., Hakimpour, P., Landgraf, P., Rice, A., Tuschl, T., Casellas, R., and
Papavasiliou, F.N. (2008). MicroRNA-155 is a negative regulator of activation-induced
cytidine deaminase. Immunity 28, 621–629.

Tonegawa, S. (1983). Somatic generation of antibody diversity. Nature 302, 575–581.

Trageser, D., Iacobucci, I., Nahar, R., Duy, C., Von Levetzow, G., Klemm, L., Park, E.,
Schuh, W., Gruber, T., Herzog, S., et al. (2009). Pre-B cell receptor-mediated cell cycle
arrest in Philadelphia chromosome-positive acute lymphoblastic leukemia requires
IKAROS function. J. Exp. Med. 206, 1739–1753.

Tsai, A.G., Lu, H., Raghavan, S.C., Muschen, M., Hsieh, C.-L., and Lieber, M.R. (2008).
Human chromosomal translocations at CpG sites and a theoretical basis for their lineage
and stage specificity. Cell 135, 1130–1142.

Türkmen, S., Riehn, M., Klopocki, E., Molkentin, M., Reinhardt, R., and Burmeister, T.
(2011). A BACH2-BCL2L1 fusion gene resulting from a t(6;20)(q15;q11.2) chromosomal
translocation in the lymphoma cell line BLUE-1. Genes Chromosomes Cancer 50, 389–
396.

Urbánek, P., Wang, Z.Q., Fetka, I., Wagner, E.F., and Busslinger, M. (1994). Complete
block of early B cell differentiation and altered patterning of the posterior midbrain in
mice lacking Pax5/BSAP. Cell 79, 901–912.

Vafa, O., Wade, M., Kern, S., Beeche, M., Pandita, T.K., Hampton, G.M., and Wahl, G.M.
(2002). c-Myc can induce DNA damage, increase reactive oxygen species, and mitigate
p53 function: a mechanism for oncogene-induced genetic instability. Mol. Cell 9, 1031–
1044.

127
Van Zelm, M.C., Van der Burg, M., De Ridder, D., Barendregt, B.H., De Haas, E.F.E.,
Reinders, M.J.T., Lankester, A.C., Révész, T., Staal, F.J.T., and Van Dongen, J.J.M.
(2005). Ig gene rearrangement steps are initiated in early human precursor B cell
subsets and correlate with specific transcription factor expression. J. Immunol. 175,
5912–5922.

Vassiliou, G.S., Cooper, J.L., Rad, R., Li, J., Rice, S., Uren, A., Rad, L., Ellis, P.,
Andrews, R., Banerjee, R., et al. (2011). Mutant nucleophosmin and cooperating
pathways drive leukemia initiation and progression in mice. Nat. Genet. 43, 470–475.

Wang, J.H., Nichogiannopoulou, A., Wu, L., Sun, L., Sharpe, A.H., Bigby, M., and
Georgopoulos, K. (1996). Selective defects in the development of the fetal and adult
lymphoid system in mice with an Ikaros null mutation. Immunity 5, 537–549.

Watanabe-Matsui, M., Muto, A., Matsui, T., Itoh-Nakadai, A., Nakajima, O., Murayama,
K., Yamamoto, M., Ikeda-Saito, M., and Igarashi, K. (2011). Heme regulates B-cell
differentiation, antibody class switch, and heme oxygenase-1 expression in B cells as a
ligand of Bach2. Blood 117, 5438–5448.

Wiemels, J.L., Ford, A.M., Van Wering, E.R., Postma, A., and Greaves, M. (1999).
Protracted and variable latency of acute lymphoblastic leukemia after TEL-AML1 gene
fusion in utero. Blood 94, 1057–1062.

Wossning, T., Herzog, S., Köhler, F., Meixlsperger, S., Kulathu, Y., Mittler, G., Abe, A.,
Fuchs, U., Borkhardt, A., and Jumaa, H. (2006). Deregulated Syk inhibits differentiation
and induces growth factor-independent proliferation of pre-B cells. J. Exp. Med. 203,
2829–2840.

Xu, Z., Pone, E.J., Al-Qahtani, A., Park, S.-R., Zan, H., and Casali, P. (2007). Regulation
of aicda expression and AID activity: Relevance to somatic hypermutation and class
switch DNA recombination. Crit Rev Immunol 27, 367–397.

Yamamoto, M., Hayashi, K., Nojima, T., Matsuzaki, Y., Kawano, Y., Karasuyama, H.,
Goitsuka, R., and Kitamura, D. (2006). BASH-novel PKC-Raf-1 pathway of pre-BCR
signaling induces kappa gene rearrangement. Blood 108, 2703–2711.

Yount, W.J., Dorner, M.M., Kunkel, H.G., and Kabat, E.A. (1968). Studies on human
antibodies. VI. Selective variations in subgroup composition and genetic markers. J. Exp.
Med. 127, 633–646.

128
Appendices
Appendix A: Supplementary information for Chapter 3
Figure A.1. Infectious and inflammatory origins of childhood pre-B ALL.
Loss of IL7R signaling at Fraction D following the completion of pre-BCR checkpoint
control (Fraction C’) makes a pre-B cell vulnerable to acquisition of genetic changes.

Figure A.2. IL7R signaling protects against leukemogenesis at the pre-BCR
checkpoint by preventing genetic instability.
Flowchart highlighting the molecular regulators of Aid and Rags during the transition
from Fraction C’ to D. IL7R safeguards against pre-mature AID and RAG expression. On
the contrary, pre-BCR and Toll-Like Receptor (TLR) oppose IL7R signaling and induce
expression of AID and RAGs.

129
Figure A.3. IL7R signaling prevents LPS responsiveness by blocking AID
expression.
A. Expression levels of Aid mRNA in Slp65
-/-
IL7-dependent pre-B cells at Fractions C’
and D, before and after treatment with LPS was measured by qRT-PCR. Differentiation
from Fraction C’ to D was carried out by IL7 withdrawal. B. Same experiment as in A, but
differentiation from Fraction C’ to D was induced by Slp65 reconstitution in Slp65
-/-
pre-B
cells. C. Western blot to compare AID protein levels at Fractions C’ and D after exposure
to LPS. Splenocytes from a wildtype mouse induced with LPS and IL4 were used as a
positive control for AID expression.

130
Figure A.4. Mouse spectral karyotyping (mSKY) reveals aneuploidy and
chromosomal translocations in TEL-AML1 pre-B ALL.
SKY was carried out using leukemic bone marrows from two mice injected with Aid
+/+

Rag1
+/+
-IL7+LPS TEL-AML1 positive cells. 18 individual cells from each mouse were
analyzed for chromosomal aberrations. Shown below are 6 representative cells which
highlight the acquisition of additional genetic alterations (denoted by white circles) after
exposure of the Fraction D pre-B cells to inflammatory agent (LPS).

131
Table A1. Sequences of oligonucleotide primers for qRT-PCR used in chapter 3

Hprt_F 5’-GGGGGCTATAAGTTCTTTGC-3’
Hprt_R 5’-TCCAACACTTCGAGAGGTCC-3’
Rag2_F 5’-GCAGATGGTAACAGTGGGTC-3’
Rag2_R 5’-ATTGCAGGCTTCAGTTTGAG-3’
Rag1_F 5’-TAACAACCAAGCTGCAGACA-3’
Rag1_R 5’-CCTCTGAGGAATCCTTCTCC-3’
Aid_F 5’- AAATGTCCGCTGGGCCAA-3’
Aid_R 5’- CATCGACTTCGTACAAGGG-3’

132
Appendix B: Supplementary information for Chapter 5
Figure B.1. Sequence analysis of V
H
(D)J
H
junctions in Bach2
+/+
and Bach2
-/-
bone
marrow and splenic B cells.
CD19
+
B cells from bone marrow and spleen of Bach2
+/+
and Bach2
-/-
mice were isolated
by MACS. RT-PCR using VJ558 and Cμ was carried out to amplify all possible IgH
rearrangements in the B cell repertoire and subjected to sequencing analysis as
described in Section 5.2.8. Sequences with ‘#’ and ‘
*
’ indicate non-productive V(D)J
rearrangements.

Figure B.2. Inducible overexpression of Bach2 aids the clearance of pre-B cells
with non-productive V
H
(D)J
H
rearrangements.
Bach2
+/+
and Bach2
-/-
IL-7 dependent pre-B cells were transduced with empty vector
(EV-ER
T2
) or with Bach2-ER
T2
. Induction with tamoxifen for 24 hours resulted in BACH2
translocation to the nucleus in the groups carrying the Bach2-ER
T2
construct. Following
this, V(D)J junctions were sequenced. Inducible overexpression of Bach2 in Bach2
-/-
pre-
B cells significantly lowered the number of non-productive heavy chain rearrangements
confirming the requirement of BACH2 in the clearance of pre-B cells with non-productive
rearrangements.

133
Figure B.2. continued.

Figure B.3. Bach2
-/-
pre-B cells display reduced expression of Bach2 mRNA.
qRT-PCR to determine the levels of Bach2 mRNA in Bach2
+/+
and Bach2
-/-
BCR-ABL1-
transformed cells before and after differentiation with imatinib (IM, n=2, each condition).

134
Figure B.4. Overexpression of Bach2 triggers expression of Rag1/ Rag2
recombinase in pre-B cells.
Quantitative RT-PCR to measure levels of Rag1/Rag2 mRNA after Bach2
overexpression.

Table B1: List of oligonucleotide primers for chapter 5
Primers for cloning MSCV Bach2-ER
T2
IRES GFP plasmid
BACH2_F 5’- AAAGGATCCGTCTGATCCCTTGCT -3’
BACH2_R 5’- AAACTCGAGGGTATAATCTTTCCT -3’

Primers for clonality and spectratyping analysis

V
H
1_F 5’- AAGGCCACACTGACTGTAGAC -3’
Cμ_R 5’- TGGCCACCAGATTCTTATCAG -3’
Cμ-FAM_R 5’- AGACGAGGGGGAAGACATTTG -3’

135
Appendix C: Supplementary information for Chapter 6
Figure C.1. BACH2 is increased in Ph
+
ALL cells upon TKI treatment.
A. Comparison of gene expression values of BACH2 in human Ph
+
cell lines before and
after imatinib (IM) treatment. B. Western blot depicting increase in BACH2 upon
treatment with a TKI like IM (10μM, 24 hours).

Figure C.2. BACH2 is affected by somatic mutations in a fraction of Ph
+
ALL cases.
Diagrammatic representation of the point mutations identified in BACH2 coding region in
10 cases of primary Ph
+
ALL.

136
Figure C.3. Bach2 prevents leukemic transformation by Myc in vivo.
Flow cytometry for CD19/Myc-GFP double positive cells in bone marrows and spleens
isolated from mice in groups injected with either Bach2+/+ or Bach2-/- Myc-GFP IL-7
dependent cells.

Figure C.4. Btg2 is downregulated upon loss of Bach2.
qRT-PCR comparing mRNA levels of Btg2 in Bach2
+/+
and Bach2
-/-
BCR-ABL1-
transformed cells in the presence and absence of TKI (imatinib) treatment (n=3, each
condition).

137
Table C1. Sequences of oligonucleotide primers used in chapter 6.
Primers used for quantitative RT-PCR
Bcl6_F 5’-CCTGCAACTGGAAGAAGTATAAG-3’
Bcl6_R 5’-AGTATGGAGGCACATCTCTGTAT-3’
Bach2_F 5’-TGAGGTACCCACAGACACCA-3’
Bach2_R 5’-TGCCAGGACTGTCTTCACTG-3’
Hprt_F 5’-GGGGGCTATAAGTTCTTTGC-3’
Hprt_R 5’-TCCAACACTTCGAGAGGTCC-3’
Btg2_F 5’-GATGGCTCCATCTGTGTCCT-3’
Btg2_R 5’-TATACGGTGGCCTGTTGTCA-3’
Rag2_F 5’-GCAGATGGTAACAGTGGGTC-3’
Rag2_R 5’-ATTGCAGGCTTCAGTTTGAG-3’
Rag1_F 5’-TAACAACCAAGCTGCAGACA-3’
Rag1_R 5’-CCTCTGAGGAATCCTTCTCC-3’
GFP_F 5’-AGGAGCGCACCATCTTCTT-3’
GFP_R 5’-GCCATGATATAGACGTTGTGG-3’
Trp53_F 5’-TCCTTACCATCATCACACTGG-3’
Trp53_R 5’-CGGATCTTGAGGGTGAAATAC-3’
Cdkn2a_F 5’-GGACCAGGTGATGATGATG-3’
Cdkn2a_R 5’-ATCGCACGATGTCTTGATG-3’

Primers used for quantitative chromatin immunoprecipitation (Q-ChIP)

CDKN2A_F 5’-TAGATGGACTCGGAGCAAGG-3’
CDKN2A_R 5’-TTTCGCTCCGGTTAACTTTC-3’
Trp53 region1_F 5’-GCCGAGGCTAGAGTGCATTA-3’
Trp53 region1_R 5’-TCCCTGGTGATTGCTTTAGG-3’
Trp53 region2_F 5’-GAAACCCTGGGGTTGATTTT-3’
Trp53 region2_R 5’-AGTTCCAGGCAAACATGGAC-3’
Bcl6 exon1_F 5’-CCGAGAATTGAGCTCTGTTGA-3’
Bcl6 exon1_R 5’-GGCAGCAACAGCAATAATCA-3’
ACTA1_F 5’-AGAGTCAGAGCAGCAGGTAG-3’
ACTA1_R 5’-CAAGGCTCAATAGCTTTCTT-3’

138
Mutation analysis in Bach2 translated region (from Bach2 cDNA)
Primers to amplify Bach2 translated region

SetA_F 5’-TTACATGGTGTGAACGGCATG-3’
SetA_R 5’-CCTGGCTGTGACCTCCTC-3’
SetB_F 5’-AGGAGGTCACAGCCAGG-3’
SetB_R 5’-GATGCTCTCTTCCTCATTCT-3’
SetC_F 5’-ACGCTCTGCCTGTCTGGAGA-3’
SetC_R 5’-CGGCTCAGAGAGGTCTTTGT-3’
SetD_F 5’-GTGCCAAAGGGTCTGTGGGT-3’
SetD_R 5’-CTCACACACCAATTTGCGGA-3’
SetE_F 5’-AAAGAGAAACTGTTGTCAGAG-3’
SetE_R 5’-CTAGGTATAATCTTTCCTGG-3’

Primers for sequencing Bach2 translated region

SetA seq_F 5’-GTGTGAACGGCATGTCTGTG-3’
SetA seq_R 5’-CCTGGCTGTGACCTCCTC-3’
SetB seq_F 5’-TCACAGCCAGGGGCTTTG -3’
SetB seq_R 5’-GATGCTCTCTTCCTCATTCT-3’
SetC seq_F 5’-GCCTGTCTGGAGATGAGCC-3’
SetC seq_R 5’-CGGCTCAGAGAGGTCTTTGT-3’
SetD seq_F 5’-GGGTCTGTGGGTGGGAGC-3’
SetD seq_R 5’-CTCACACACCAATTTGCGGA-3’
SetE seq_F 5’-AAACTGTTGTCAGAGAGGAAT-3’
SetE seq_R 5’-CTAGGTATAATCTTTCCTGG-3’

*Position of primers used for mutation analysis on Bach2 cDNA

Name of primer Region of Bach2 cDNA spanned (bp)
SetA 691 - 963
SetB 947 - 1575
SetC 1576 - 2091
SetD 2092 - 2760
SetE 2761 - 3234
SetA seq 698 - 963
SetB seq 953 - 1575
SetC seq 1583 - 2091
SetD seq 2100 - 2760
SetE seq 2767 - 3234

139
Table C2. Somatic mutations of the BACH2 gene in patient-derived Ph
+
ALL cells.

Ph
+
ALL
case
N Mutation Amino
Acid
change
Disease Karyotype
TXL2 0 at diagnosis t(9;22)(q34;q11)
TXL3 1 C1039T R111C at diagnosis t(9;22)(q34;q11)
ICN1 2
C1779T
C1039T
Silent
R111C
at diagnosis t(9;22)(q34;q11)
SFO2 1 C1039T R111C at diagnosis t(9;22)(q34;q11)
LAX2 0
T315I, Relapse
(Imatinib)
t(9;22)(q34;q11)
LAX9 1 C1779T Silent at diagnosis t(9;22)(q34;q11)
BLQ1 2
G938T
C1039T
S77I
R111C
T315I, Relapse
(Imatinib)
FISH der(9), der(22)
BLQ5 1 C974T P89L
T315I, Relapse
(Imatinib)
FISH der(9), der(22)
BLQ11 0
T315I, Relapse
(Imatinib)
FISH der(9), der(22)
PDX59 2
A2214G
C1039T
Silent
R111C

46,XY,del(9)(p13),t(9;2
2)(q34;q11.2)[12]/46,X
Y,del(9)(p13),der(9)t(9;
22)(q34;q11.2),ider(22)
(q10)t(9;22)(q34;q11.2)
[4]/46,XY[7]

Abstract (if available)

Abstract Leukemogenesis is a multi-step process which involves the disruption of the normal development of a hematopoetic cell. Precursor-B acute lymphoblastic leukemia (Pre-B ALL) is one such example of the process of leukemogenesis that results from deregulated early B cell development in the bone marrow. Typically, checkpoints which are present at every stage of development safeguard an early B cell from leukemic transformation. This thesis focuses on two such critical cell signaling mechanisms that act at the pre-B cell receptor (pre-BCR) checkpoint (Fraction C’), and thereby protect the cell from overt leukemogenesis. While one of the mechanisms protects the B cell from acquiring genetic alterations, the other eliminates deleterious B cells by a process known as negative selection. ❧ In the first half of this thesis, we highlight the novel role of IL7R as the guardian of the B cell genome before Fraction D. We show that IL7R carries out this function by preventing pre-mature expression of AID and deleterious rise in the levels of the RAG1/RAG2 recombinases. Both these enzymes are notorious in causing genetic instability and have been previously implicated in the process of leukemogenesis. However, the natural safeguard mechanisms that prevent their concomitant expression have not been elucidated. Our study provides novel insight in this direction. In addition, we shed light on the role of inflammatory cues in accelerating the process of leukemogenesis, when the safeguard is lost or compromised. We demonstrate that this study is particularly relevant to the clonal evolution of certain childhood leukemia like the TEL-AML1 subgroup, which require second hits for full-blown leukemogenesis to occur. ❧ In the latter half of this thesis, we focus on another mechanism of leukemic transformation that occurs at the pre-BCR checkpoint. We show that the natural process of negative selection (apoptosis) at the pre-BCR checkpoint is crucial to thwart leukemogenesis. We identify the BTB domain containing B-lymphoid protein BACH2 as the key inducer of the negative selection process. Role of BACH2 has not been well-characterized in early B cells. Therefore, our study is the first of its kind to highlight the role of BACH2 in negative selection. By extrapolating our understanding of BACH2 in B cell development to pre-B ALL, we demonstrate that this protein possesses leukemia-suppressive properties. Therefore, BACH2 represents another paradigm of safeguard against leukemogenesis at the pre-BCR checkpoint. ❧ Through our studies on both IL7R and BACH2, we highlight the importance of a tightly controlled early B cell development process that safeguards the cell from leukemic transformation. Such an understanding of leukemogenesis is required for identifying better treatment strategies for ALL patients.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Induction of hypersignaling as a therapeutic approach for treatment of BCR-ABL1 positive Acute Lymphoblastic Leukemia (ALL) cells

PDF

The role of survivin in drug resistant pediatric acute lymphoblastic leukemia

PDF

Pre-B cell receptor signaling in acute lymphoblastic leukemia

PDF

The role of adipocytes in acute lymphoblastic leukemia cell migration and survival against daunorubicin

PDF

TNF-alpha, an early hepatocellular response to HCV infection

PDF

Therapeutic resistance in acute lymphoblastic leukemia: cIAP2 inhibition sensitizes B cell acute lymphoblastic leukemia to anti-CD19 chimeric antigen receptor T cell

PDF

Understanding the role of BRCA1 and Aurora A kinase in polyploidy development in ovarian carcinoma precursor cells

PDF

Role of integrin α4 in drug resistant acute lymphoblastic leukemia

PDF

The essential role of histone H2A deubiquitinase MYSM1 in natural killer cell maturation and HSC homeostasis

PDF

Mechanisms of nucleases in non-homologous DNA end joining

PDF

Role of the bone marrow niche components in B cell malignancies

PDF

Understanding acute lymphoblastic leukemia in different ethnic groups in the United States

PDF

Kaposi's sarcoma associated herpes-virus induces cellular proxi expression to modulate host gene expression that benefits viral infection and oncogenesis

PDF

The cancer stem-like phenotype: therapeutics, phenotypic plasticity and mechanistic studies

PDF

Tissue interactions & molecular pathways in specification of the ectomesenchyme from cranial neural crest

PDF

The role of endoplasmic reticulum protein GRP78 in normal hematopoeises and PTEN-null leukemogenesis

PDF

The function of BS69 in mouse embryogenesis and embryonic stem cell differentiation

PDF

Antibody discovery and structural studies of the viral oncogene ORF74

PDF

The mechanisms of somatic cell reprogramming

PDF

Investigation of the molecular mechanisms of ototoxicity

Asset Metadata

Creator Swaminathan, Srividya (author)

Core Title Loss of checkpoint controls in acute lymphoblastic leukemia

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Genetic, Molecular and Cellular Biology

Publication Date 10/03/2013

Defense Date 03/11/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag AID,BACH2,OAI-PMH Harvest,PAX5,pre-BCR checkpoint

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Muschen, Markus (committee chair), Heisterkamp, Eleanora C. (committee member), Jung, Jae U. (committee member), Kim, Yong-Mi (committee member), Lieber, Michael R. (committee member)

Creator Email srividys@usc.edu,vid1787@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-231464

Unique identifier UC11295103

Identifier usctheses-c3-231464 (legacy record id)

Legacy Identifier etd-Swaminatha-1510.pdf

Dmrecord 231464

Document Type Dissertation

Rights Swaminathan, Srividya

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA